source: other-projects/nightly-tasks/diffcol/trunk/model-collect/Word-PDF-Formatting/archives/HASHeaa2.dir/doc.xml

Last change on this file was 37437, checked in by anupama, 14 months ago

AUTOCOMMIT by gen-model-colls.sh script. Message: Clean rebuild of model collections 1/2. Clearing out deprecated archives and index.
File size: 246.0 KB
Line 
1<?xml version="1.0" encoding="utf-8" standalone="no"?>
2<!DOCTYPE Archive SYSTEM "https://greenstone.org/dtd/Archive/1.0/Archive.dtd">
3<Archive>
4<Section>
5 <Description>
6 <Metadata name="gsdldoctype">indexed_doc</Metadata>
7 <Metadata name="Language">en</Metadata>
8 <Metadata name="Encoding">utf8</Metadata>
9 <Metadata name="GENERATOR">wvWare/wvWare version 1.2.4</Metadata>
10 <Metadata name="Title">1997-00 Listing of Working Papers</Metadata>
11 <Metadata name="URL">http://Scratch/ak19/gs2-diffcol-7Mar2023/collect/Word-PDF-Formatting/tmp/1678163165_3/word01.html</Metadata>
12 <Metadata name="UTF8URL">http://Scratch/ak19/gs2-diffcol-7Mar2023/collect/Word-PDF-Formatting/tmp/1678163165_3/word01.html</Metadata>
13 <Metadata name="gsdlsourcefilename">import/word01.doc</Metadata>
14 <Metadata name="gsdlsourcefilerenamemethod">url</Metadata>
15 <Metadata name="gsdlconvertedfilename">tmp/1678163165_3/word01.html</Metadata>
16 <Metadata name="OrigSource">word01.html</Metadata>
17 <Metadata name="Source">word01.doc</Metadata>
18 <Metadata name="SourceFile">word01.doc</Metadata>
19 <Metadata name="Plugin">WordPlugin</Metadata>
20 <Metadata name="FileSize">110080</Metadata>
21 <Metadata name="FilenameRoot">word01</Metadata>
22 <Metadata name="FileFormat">Word</Metadata>
23 <Metadata name="srcicon">_icondoc_</Metadata>
24 <Metadata name="srclink_file">doc.doc</Metadata>
25 <Metadata name="srclinkFile">doc.doc</Metadata>
26 <Metadata name="Identifier">HASHeaa2992e081949673150f3</Metadata>
27 <Metadata name="lastmodified">1678162980</Metadata>
28 <Metadata name="lastmodifieddate">20230307</Metadata>
29 <Metadata name="oailastmodified">1678163165</Metadata>
30 <Metadata name="oailastmodifieddate">20230307</Metadata>
31 <Metadata name="assocfilepath">HASHeaa2.dir</Metadata>
32 <Metadata name="gsdlassocfile">doc.doc:application/msword:</Metadata>
33 </Description>
34 <Content>
35
36&lt;!--Section Begins--&gt;&lt;br&gt;
37
38
39
40&lt;p&gt;&lt;div name=&quot;Title&quot; align=&quot;center&quot; style=&quot;margin: 0.00mm -0.12mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
41
42&lt;p style=&quot;text-indent: 0.00mm; text-align: center; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
43&lt;b&gt;1997-00 Listing of Working Papers &lt;/b&gt;
44&lt;/p&gt;&lt;/div&gt;
45
46
47
48&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
49
50&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
51
52&lt;/p&gt;&lt;/div&gt;
53
54
55
56&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
57
58&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
59
60&lt;/p&gt;&lt;/div&gt;
61
62
63
64&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
65
66&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
67
68&lt;/p&gt;&lt;/div&gt;
69
70
71
72&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
73
74&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
752000/1
76&lt;/p&gt;&lt;/div&gt;
77
78
79
80&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
81
82&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
83Using compression to identify acronyms in text
84&lt;/p&gt;&lt;/div&gt;
85
86
87
88&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
89
90&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
91Stuart Yeates, David Bainbridge, Ian H. Witten
92&lt;/p&gt;&lt;/div&gt;
93
94
95
96&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
97
98&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
99Text mining is about looking for patterns in natural language text, and may be defined as the process of analyzing text to extract information from it for particular purposes. In previous work, we claimed that compression is a key technology for text mining, and backed this up with a study that showed how particular kinds of lexical tokens&amp;mdash;names, dates, locations, &lt;i&gt;etc.&lt;/i&gt;&amp;mdash;can be identified and located in running text, using compression models to provide the leverage necessary to distinguish different token types (Witten &lt;i&gt;et al.&lt;/i&gt;, 1999)
100&lt;/p&gt;&lt;/div&gt;
101
102
103
104&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
105
106&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
107
108&lt;/p&gt;&lt;/div&gt;
109
110
111
112&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
113
114&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1152000/2
116&lt;/p&gt;&lt;/div&gt;
117
118
119
120&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
121
122&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
123Text categorization using compression models
124&lt;/p&gt;&lt;/div&gt;
125
126
127
128&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
129
130&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
131Eibe Frank, Chang Chui, Ian H. Witten
132&lt;/p&gt;&lt;/div&gt;
133
134
135
136&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
137
138&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
139Text categorization, or the assignment of natural language texts to predefined categories based on their content, is of growing importance as the volume of information available on the internet continues to overwhelm us. The use of predefined categories implies a &amp;ldquo;supervised learning&amp;rdquo; approach to categorization, where already-classified articles - which effectively define the categories - are used as &amp;ldquo;training data&amp;rdquo; to build a model that can be used for classifying new articles that comprise the &amp;ldquo;test data&amp;rdquo;. This contrasts with &amp;ldquo;unsupervised&amp;rdquo; learning, where there is no training data and clusters of like documents are sought amongst the test articles. With supervised learning, meaningful labels (such as keyphrases) are attached to the training documents, and appropriate labels can be assigned automatically to test documents depending on which category they fall into.
140&lt;/p&gt;&lt;/div&gt;
141
142
143
144&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
145
146&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
147
148&lt;/p&gt;&lt;/div&gt;
149
150
151
152&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
153
154&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1552000/3
156&lt;/p&gt;&lt;/div&gt;
157
158
159
160&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
161
162&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
163Reserved for Sally Jo
164&lt;/p&gt;&lt;/div&gt;
165
166
167
168&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
169
170&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
171
172&lt;/p&gt;&lt;/div&gt;
173
174
175
176&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
177
178&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1792000/4
180&lt;/p&gt;&lt;/div&gt;
181
182
183
184&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
185
186&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
187Interactive machine learning&amp;mdash;letting users build classifiers
188&lt;/p&gt;&lt;/div&gt;
189
190
191
192&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
193
194&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
195Malcolm Ware, Eibe Frank, Geoffrey Holmes, Mark Hall, Ian H. Witten
196&lt;/p&gt;&lt;/div&gt;
197
198
199
200&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
201
202&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
203According to standard procedure, building a classifier is a fully automated process that follows data preparation by a domain expert. In contrast, &amp;lt;I&amp;gt;interactive&amp;lt;/I&amp;gt;machine learning engages users in actually generating the classifier themselves. This offers a natural way of integrating background knowledge into the modeling stage&amp;mdash;so long as interactive tools can be designed that support efficient and effective communication. This paper shows that appropriate techniques can empower users to create models that compete with classifiers built by state-of-the-art learning algorithms. It demonstrates that users&amp;mdash;even users who are not domain experts&amp;mdash;can often construct good classifiers, without any help from a learning algorithm, using a simple two-dimensional visual interface. Experiments demonstrate that, not surprisingly, success hinges on the domain: if a few attributes can support good predictions, users generate accurate classifiers, whereas domains with many high-order attribute interactions favor standard machine learning techniques. The future challenge is to achieve a symbiosis between human user and machine learning algorithm.
204&lt;/p&gt;&lt;/div&gt;
205
206
207
208&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
209
210&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
211
212&lt;/p&gt;&lt;/div&gt;
213
214
215
216&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
217
218&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2192000/5
220&lt;/p&gt;&lt;/div&gt;
221
222
223
224&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
225
226&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
227KEA: Practical automatic keyphrase extraction
228&lt;/p&gt;&lt;/div&gt;
229
230
231
232&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
233
234&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
235Ian H. Witten, Gordon W. Paynter, Eibe Frank, Carl Gutwin, Craig G. Nevill-Manning
236&lt;/p&gt;&lt;/div&gt;
237
238
239
240&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
241
242&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
243Keyphrases provide semantic metadata that summarize and characterize documents. This paper describes Kea, an algorithm for automatically extracting keyphrases from text. Kea identifies candidate keyphrases using lexical methods, calculates feature values for each candidate, and uses a machine learning algorithm to predict which candidates are good keyphrases. The machine learning scheme first builds a prediction model using training documents with known keyphrases, and then uses the model to find keyphrases in new documents. We use a large test corpus to evaluate Kea's effectiveness in terms of how many author-assigned keyphrases are correctly identified. The system is simple, robust, and publicly available.
244&lt;/p&gt;&lt;/div&gt;
245
246
247
248&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
249
250&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
251
252&lt;/p&gt;&lt;/div&gt;
253
254
255
256&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
257
258&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2592000/6
260&lt;/p&gt;&lt;/div&gt;
261
262
263
264&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
265
266&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
267&lt;i&gt;Ό&lt;/i&gt;-Charts and Z: hows, whys and wherefores
268&lt;/p&gt;&lt;/div&gt;
269
270
271
272&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
273
274&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
275Greg Reeve, Steve Reeves
276&lt;/p&gt;&lt;/div&gt;
277
278
279
280&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
281
282&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
283In this paper we show, by a series of examples, how the &lt;i&gt;Ό&lt;/i&gt;-chart formalism can be translated into Z. We give reasons for why this is an interesting and sensible thing to do and what it might be used for.
284&lt;/p&gt;&lt;/div&gt;
285
286
287
288&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
289
290&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
291
292&lt;/p&gt;&lt;/div&gt;
293
294
295
296&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
297
298&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
299
300&lt;/p&gt;&lt;/div&gt;
301
302
303
304&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
305
306&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3072000/7
308&lt;/p&gt;&lt;/div&gt;
309
310
311
312&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
313
314&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
315One dimensional non-uniform rational B-splines for animation control
316&lt;/p&gt;&lt;/div&gt;
317
318
319
320&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
321
322&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
323Abdelaziz Mahoui
324&lt;/p&gt;&lt;/div&gt;
325
326
327
328&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
329
330&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
331Most 3D animation packages use graphical representations called motion graphs to represent the variation in time of the motion parameters. Many use two-dimensional B-splines as animation curves because of their power to represent free-form curves. In this project, we investigate the possibility of using One-dimensional Non-Uniform Rational B-Spline (NURBS) curves for the interactive construction of animation control curves. One-dimensional NURBS curves present the potential of solving some problems encountered in motion graphs when two-dimensional B-splines are used. The study focuses on the properties of One-dimensional NURBS mathematical model. It also investigates the algorithms and shape modification tools devised for two-dimensional curves and their port to the One-dimensional NURBS model. It also looks at the issues related to the user interface used to interactively modify the shape of the curves.
332&lt;/p&gt;&lt;/div&gt;
333
334
335
336&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
337
338&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
339
340&lt;/p&gt;&lt;/div&gt;
341
342
343
344&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
345
346&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3472000/8
348&lt;/p&gt;&lt;/div&gt;
349
350
351
352&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
353
354&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
355Correlation-based feature selection of discrete and numeric class machine learning
356&lt;/p&gt;&lt;/div&gt;
357
358
359
360&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
361
362&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
363Mark A. Hall
364&lt;/p&gt;&lt;/div&gt;
365
366
367
368&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
369
370&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
371Algorithms for feature selection fall into two broad categories: &amp;lt;I&amp;gt;wrappers&amp;lt;/I&amp;gt;that use the learning algorithm itself to evaluate the usefulness of features and &amp;lt;I&amp;gt;filters&amp;lt;/I&amp;gt;that evaluate features according to heuristics based on general characteristics of the data. For application to large databases, filters have proven to be more practical than wrappers because they are much faster. However, most existing filter algorithms only work with discrete classification problems. This paper describes a fast, correlation-based filter algorithm that can be applied to continuous and discrete problems. The algorithm often out-performs the well-known ReliefF attribute estimator when used as a preprocessing step for naïve Bayes, instance-based learning, decision trees, locally weighted regression, and model trees. It performs more feature selection than ReliefF does-reducing the data dimensionality by fifty percent in most cases. Also, decision and model trees built from the prepocessed data are often significantly smaller.
372&lt;/p&gt;&lt;/div&gt;
373
374
375
376&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
377
378&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
379
380&lt;/p&gt;&lt;/div&gt;
381
382
383
384&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
385
386&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3872000/9
388&lt;/p&gt;&lt;/div&gt;
389
390
391
392&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
393
394&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
395A development environment for predictive modelling in foods
396&lt;/p&gt;&lt;/div&gt;
397
398
399
400&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
401
402&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
403G. Holmes, M.A. Hall
404&lt;/p&gt;&lt;/div&gt;
405
406
407
408&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
409
410&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
411WEKA (Waikato Environment for Knowledge Analysis) is a comprehensive suite of Java class libraries that implement many state-of-the-art machine learning/data mining algorithms. Non-programmers interact with the software via a user interface component called the Knowledge Explorer.
412&lt;/p&gt;&lt;/div&gt;
413
414
415
416&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
417
418&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
419
420&lt;/p&gt;&lt;/div&gt;
421
422
423
424&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
425
426&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
427Applications constructed from the WEKA class libraries can be run on any computer with a web browsing capability, allowing users to apply machine learning techniques to their own data regardless of computer platform. This paper describes the user interface component of the WEKA system in reference to previous applications in the predictive modeling of foods.
428&lt;/p&gt;&lt;/div&gt;
429
430
431
432&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
433
434&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
435
436&lt;/p&gt;&lt;/div&gt;
437
438
439
440&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
441
442&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4432000/10
444&lt;/p&gt;&lt;/div&gt;
445
446
447
448&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
449
450&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
451Benchmarking attribute selection techniques for data mining
452&lt;/p&gt;&lt;/div&gt;
453
454
455
456&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
457
458&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
459Mark A. Hall, Geoffrey Holmes
460&lt;/p&gt;&lt;/div&gt;
461
462
463
464&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
465
466&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
467Data engineering is generally considered to be a central issue in the development of data mining applications. The success of many learning schemes, in their attempts to construct models of data, hinges on the reliable identification of a small set of highly predictive attributes. The inclusion of irrelevant, redundant and noisy attributes in the model building process phase can result in poor predictive performance and increased computation.
468&lt;/p&gt;&lt;/div&gt;
469
470
471
472&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
473
474&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
475
476&lt;/p&gt;&lt;/div&gt;
477
478
479
480&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
481
482&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
483Attribute selection generally involves a combination of search and attribute utility estimation plus evaluation with respect to specific learning schemes. This leads to a large number of possible permutations and has led to a situation where very few benchmark studies have been conducted.
484&lt;/p&gt;&lt;/div&gt;
485
486
487
488&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
489
490&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
491
492&lt;/p&gt;&lt;/div&gt;
493
494
495
496&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
497
498&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
499This paper presents a benchmark comparison of several attribute selection methods. All the methods produce an attribute ranking, a useful devise of isolating the individual merit of an attribute. Attribute selection is achieved by cross-validating the rankings with respect to a learning scheme to find the best attributes. Results are reported for a selection of standard data sets and two learning schemes C4.5 and naïve Bayes.
500&lt;/p&gt;&lt;/div&gt;
501
502
503
504&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
505
506&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
507
508&lt;/p&gt;&lt;/div&gt;
509
510
511
512&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
513
514&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
5152000/11
516&lt;/p&gt;&lt;/div&gt;
517
518
519
520&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
521
522&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
523Steve Reeves, Greg Reeve
524&lt;/p&gt;&lt;/div&gt;
525
526
527
528&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
529
530&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
531
532&lt;/p&gt;&lt;/div&gt;
533
534
535
536&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
537
538&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
539
540&lt;/p&gt;&lt;/div&gt;
541
542
543
544&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
545
546&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
5472000/12
548&lt;/p&gt;&lt;/div&gt;
549
550
551
552&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
553
554&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
555Malika Mahoui, Sally Jo Cunningham
556&lt;/p&gt;&lt;/div&gt;
557
558
559
560&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
561
562&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
563Transaction logs are invaluable sources of fine-grained information about users' search behavior. This paper compares the searching behavior of users across two WWW-accessible digital libraries: the New Zealand Digital Library's Computer Science Technical Reports collection (CSTR), and the Karlsruhe Computer Science Bibliographies (CSBIB) collection. Since the two collections are designed to support the same type of users-researchers/students in computer science a comparative log analysis is likely to uncover common searching preferences for that user group. The two collections differ in their content, however; the CSTR indexes a full text collection, while the CSBIB is primarily a bibliographic database. Differences in searching behavior between the two systems may indicate the effect of differing search facilities and content type.
564&lt;/p&gt;&lt;/div&gt;
565
566
567
568&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
569
570&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
571
572&lt;/p&gt;&lt;/div&gt;
573
574
575
576&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
577
578&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
579
580&lt;/p&gt;&lt;/div&gt;
581
582
583
584&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
585
586&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
587
588&lt;/p&gt;&lt;/div&gt;
589
590
591
592&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
593
594&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
595
596&lt;/p&gt;&lt;/div&gt;
597
598
599
600&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
601
602&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
603
604&lt;/p&gt;&lt;/div&gt;
605
606
607
608&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
609
610&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
611
612&lt;/p&gt;&lt;/div&gt;
613
614
615
616&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
617
618&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
61999/1
620&lt;/p&gt;&lt;/div&gt;
621
622
623
624&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
625
626&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
627Lexical attraction for text compression
628&lt;/p&gt;&lt;/div&gt;
629
630
631
632&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
633
634&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
635Joscha Bach, Ian H. Witten
636&lt;/p&gt;&lt;/div&gt;
637
638
639
640&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
641
642&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
643New methods of acquiring structural information in text documents may support better compression by identifying an appropriate prediction context for each symbol. The method of &amp;ldquo;lexical attraction&amp;rdquo; infers syntactic dependency structures from statistical analysis of large corpora. We describe the generation of a lexical attraction model, discuss its application to text compression, and explore its potential to outperform fixed-context models such as word-level PPM. Perhaps the most exciting aspect of this work is the prospect of using compression as a metric for structure discovery in text.
644&lt;/p&gt;&lt;/div&gt;
645
646
647
648&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
649
650&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
651
652&lt;/p&gt;&lt;/div&gt;
653
654
655
656&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
657
658&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
659
660&lt;/p&gt;&lt;/div&gt;
661
662
663
664&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
665
666&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
66799/2
668&lt;/p&gt;&lt;/div&gt;
669
670
671
672&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
673
674&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
675Generating rule sets from model trees
676&lt;/p&gt;&lt;/div&gt;
677
678
679
680&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
681
682&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
683Geoffrey Holmes, Mark Hall, Eibe Frank
684&lt;/p&gt;&lt;/div&gt;
685
686
687
688&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
689
690&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
691Knowledge discovered in a database must be represented in a form that is easy to understand. Small, easy to interpret nuggets of knowledge from data are one requirement and the ability to induce them from a variety of data sources is a second. The literature is abound with classification algorithms, and in recent years with algorithms for time sequence analysis, but relatively little has been published on extracting meaningful information from problems involving continuous classes (regression).
692&lt;/p&gt;&lt;/div&gt;
693
694
695
696&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
697
698&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
699
700&lt;/p&gt;&lt;/div&gt;
701
702
703
704&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
705
706&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
707Model trees-decision trees with linear models at the leaf nodes-have recently emerged as an accurate method for numeric prediction that produces understandable models. However, it is well known that decision lists-ordered sets of If-Then rules-have the potential to be more compact and therefore more understandable than their tree counterparts.
708&lt;/p&gt;&lt;/div&gt;
709
710
711
712&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
713
714&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
715
716&lt;/p&gt;&lt;/div&gt;
717
718
719
720&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
721
722&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
723In this paper we present an algorithm for inducing simple, yet accurate rule sets from model trees. The algorithm works by repeatedly building model trees and selecting the best rule at each iteration. It produces rule sets that are, on the whole, as accurate but smaller than the model tree constructed from the entire dataset. Experimental results for various heuristics which attempt to find a compromise between rule accuracy and rule coverage are reported. We also show empirically that our method produces more accurate and smaller rule sets than the commercial state-of-the-art rule learning system Cubist.
724&lt;/p&gt;&lt;/div&gt;
725
726
727
728&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
729
730&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
731
732&lt;/p&gt;&lt;/div&gt;
733
734
735
736&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
737
738&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
739
740&lt;/p&gt;&lt;/div&gt;
741
742
743
744&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
745
746&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
74799/3
748&lt;/p&gt;&lt;/div&gt;
749
750
751
752&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
753
754&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
755A diagnostic tool for tree based supervised classification learning algorithms
756&lt;/p&gt;&lt;/div&gt;
757
758
759
760&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
761
762&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
763Leonard Trigg, Geoffrey Holmes
764&lt;/p&gt;&lt;/div&gt;
765
766
767
768&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
769
770&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
771The process of developing applications of machine learning and data mining that employ supervised classification algorithms includes the important step of knowledge verification. Interpretable output is presented to a user so that they can verify that the knowledge contained in the output makes sense for the given application. As the development of an application is an iterative process it is quite likely that a user would wish to compare models constructed at various times or stages.
772&lt;/p&gt;&lt;/div&gt;
773
774
775
776&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
777
778&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
779
780&lt;/p&gt;&lt;/div&gt;
781
782
783
784&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
785
786&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
787One crucial stage where comparison of models is important is when the accuracy of a model is being estimated, typically using some form of cross-validation. This stage is used to establish an estimate of how well a model will perform on unseen data. This is vital information to present to a user, but it is also important to show the degree of variation between models obtained from the entire dataset and models obtained during cross-validation. In this way it can be verified that the cross-validation models are at least structurally aligned with the model garnered from the entire dataset.
788&lt;/p&gt;&lt;/div&gt;
789
790
791
792&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
793
794&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
795
796&lt;/p&gt;&lt;/div&gt;
797
798
799
800&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
801
802&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
803This paper presents a diagnostic tool for the comparison of tree-based supervised classification models. The method is adapted from work on approximate tree matching and applied to decision trees. The tool is described together with experimental results on standard datasets.
804&lt;/p&gt;&lt;/div&gt;
805
806
807
808&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
809
810&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
811
812&lt;/p&gt;&lt;/div&gt;
813
814
815
816&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
817
818&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
819
820&lt;/p&gt;&lt;/div&gt;
821
822
823
824&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
825
826&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
82799/4
828&lt;/p&gt;&lt;/div&gt;
829
830
831
832&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
833
834&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
835Feature selection for discrete and numeric class machine learning
836&lt;/p&gt;&lt;/div&gt;
837
838
839
840&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
841
842&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
843Mark A. Hall
844&lt;/p&gt;&lt;/div&gt;
845
846
847
848&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
849
850&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
851Algorithms for feature selection fall into two broad categories: &amp;lt;I&amp;gt;wrappers&amp;lt;/I&amp;gt;use the learning algorithm itself to evaluate the usefulness of features, while &amp;lt;I&amp;gt;filters&amp;lt;/I&amp;gt;evaluate features according to heuristics based on general characteristics of the data. For application to large databases, filters have proven to be more practical than wrappers because they are much faster. However, most existing filter algorithms only work with discrete classification problems.
852&lt;/p&gt;&lt;/div&gt;
853
854
855
856&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
857
858&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
859
860&lt;/p&gt;&lt;/div&gt;
861
862
863
864&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
865
866&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
867This paper describes a fast, correlation-based filter algorithm that can be applied to continuous and discrete problems. Experiments using the new method as a preprocessing step for naïve Bayes, instance-based learning, decision trees, locally weighted regression, and model trees show it to be an effective feature selector- it reduces the data in dimensionality by more than sixty percent in most cases without negatively affecting accuracy. Also, decision and model trees built from the pre-processed data are often significantly smaller.
868&lt;/p&gt;&lt;/div&gt;
869
870
871
872&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
873
874&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
875
876&lt;/p&gt;&lt;/div&gt;
877
878
879
880&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
881
882&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
883
884&lt;/p&gt;&lt;/div&gt;
885
886
887
888&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
889
890&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
89199/5
892&lt;/p&gt;&lt;/div&gt;
893
894
895
896&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
897
898&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
899Browsing tree structures
900&lt;/p&gt;&lt;/div&gt;
901
902
903
904&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
905
906&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
907Mark Apperley, Robert Spence, Stephen Hodge, Michael Chester
908&lt;/p&gt;&lt;/div&gt;
909
910
911
912&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
913
914&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
915Graphic representations of tree structures are notoriously difficult to create, display, and interpret, particularly when the volume of information they contain, and hence the number of nodes, is large. The problem of interactively browsing information held in tree structures is examined, and the implementation of an innovative tree browser described. This browser is based on distortion-oriented display techniques and intuitive direct manipulation interaction. The tree layout is automatically generated, but the location and extent of detail shown is controlled by the user. It is suggested that these techniques could be extended to the browsing of more general networks.
916&lt;/p&gt;&lt;/div&gt;
917
918
919
920&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
921
922&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
923
924&lt;/p&gt;&lt;/div&gt;
925
926
927
928&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
929
930&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
93199/6
932&lt;/p&gt;&lt;/div&gt;
933
934
935
936&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
937
938&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
939Facilitating multiple copy/past operations
940&lt;/p&gt;&lt;/div&gt;
941
942
943
944&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
945
946&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
947Mark Apperley, Jay Baker, Dale Fletcher, Bill Rogers
948&lt;/p&gt;&lt;/div&gt;
949
950
951
952&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
953
954&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
955Copy and paste, or cut and paste, using a clipboard or paste buffer has long been the principle facility provided to users for transferring data between and within GUI applications. We argue that this mechanism can be clumsy in circumstances where several pieces of information must be moved systematically. In two situations - extraction of data fields from unstructured data found in a directed search process, and reorganisation of computer program source text - we present alternative, more natural, user interface facilities to make the task less onerous, and to provide improved visual feedback during the operation.
956&lt;/p&gt;&lt;/div&gt;
957
958
959
960&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
961
962&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
963
964&lt;/p&gt;&lt;/div&gt;
965
966
967
968&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
969
970&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
971For the data extraction task we introduce the Stretchable Selection Tool, a semi-transparent overlay augmenting the mouse pointer to automate paste operations and provide information to prompt the user. We describe a prototype implementation that functions in a collaborative software environment, allowing users to cooperate on a multiple copy/paste operation. For text reorganisation, we present an extension to Emacs, providing similar functionality, but without the collaborative features.
972&lt;/p&gt;&lt;/div&gt;
973
974
975
976&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
977
978&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
979
980&lt;/p&gt;&lt;/div&gt;
981
982
983
984&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
985
986&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
98799/7
988&lt;/p&gt;&lt;/div&gt;
989
990
991
992&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
993
994&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
995Automating iterative tasks with programming by demonstration: a user evaluation
996&lt;/p&gt;&lt;/div&gt;
997
998
999
1000&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1001
1002&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1003Gordon W. Paynter, Ian H. Witten
1004&lt;/p&gt;&lt;/div&gt;
1005
1006
1007
1008&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1009
1010&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1011Computer users often face iterative tasks that cannot be automated using the tools and aggregation techniques provided by their application program: they end up performing the iteration by hand, repeating user interface actions over and over again. We have implemented an agent, called Familiar, that can be taught to perform iterative tasks using programming by demonstration (PBD). Unlike other PBD systems, it is domain independent and works with unmodified, widely-used, applications in a popular operating system. In a formal evaluation, we found that users quickly learned to use the agent to automate iterative tasks. Generally, the participants preferred to use multiple selection where possible, but could and did use PBD in situations involving iteration over many commands, or when other techniques were unavailable.
1012&lt;/p&gt;&lt;/div&gt;
1013
1014
1015
1016&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1017
1018&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1019
1020&lt;/p&gt;&lt;/div&gt;
1021
1022
1023
1024&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1025
1026&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
102799/8
1028&lt;/p&gt;&lt;/div&gt;
1029
1030
1031
1032&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1033
1034&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1035A survey of software requirements specification practices in the New Zealand software industry
1036&lt;/p&gt;&lt;/div&gt;
1037
1038
1039
1040&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1041
1042&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1043Lindsay Groves, Ray Nickson, Greg Reeve, Steve Reeves, Mark Utting
1044&lt;/p&gt;&lt;/div&gt;
1045
1046
1047
1048&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1049
1050&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1051We report on the software development techniques used in the New Zealand software industry, paying particular attention to requirements gathering. We surveyed a selection of software companies with a general questionnaire and then conducted in-depth interviews with four companies. Our results show a wide variety in the kinds of companies undertaking software development, employing a wide range of software development techniques. Although our data are not sufficiently detailed to draw statistically significant conclusions, it appears that larger software development groups typically have more well-defined software development processes, spend proportionally more time on requirements gathering, and follow more rigorous testing regimes.
1052&lt;/p&gt;&lt;/div&gt;
1053
1054
1055
1056&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1057
1058&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1059
1060&lt;/p&gt;&lt;/div&gt;
1061
1062
1063
1064&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1065
1066&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
106799/9
1068&lt;/p&gt;&lt;/div&gt;
1069
1070
1071
1072&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1073
1074&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1075The LRU*WWW proxy cache document replacement algorithm
1076&lt;/p&gt;&lt;/div&gt;
1077
1078
1079
1080&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1081
1082&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1083Chung-yi Chang, Tony McGregor, Geoffrey Holmes
1084&lt;/p&gt;&lt;/div&gt;
1085
1086
1087
1088&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1089
1090&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1091Obtaining good performance from WWW proxy caches is critically dependent on the document replacement policy used by the proxy. This paper validates the work of other authors by reproducing their studies of proxy cache document replacement algorithms. From this basis a cross-trace study is mounted. This demonstrates that the performance of most document replacement algorithms is dependent on the type of workload that they are presented with. Finally we propose a new algorithm, LRU*, that consistently performs well across all our traces.
1092&lt;/p&gt;&lt;/div&gt;
1093
1094
1095
1096&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1097
1098&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1099
1100&lt;/p&gt;&lt;/div&gt;
1101
1102
1103
1104&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1105
1106&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
110799/10
1108&lt;/p&gt;&lt;/div&gt;
1109
1110
1111
1112&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1113
1114&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1115Reduced-error pruning with significance tests
1116&lt;/p&gt;&lt;/div&gt;
1117
1118
1119
1120&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1121
1122&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1123Eibe Frank, Ian H. Witten
1124&lt;/p&gt;&lt;/div&gt;
1125
1126
1127
1128&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1129
1130&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1131When building classification models, it is common practice to prune them to counter spurious effects of the training data: this often improves performance and reduces model size. &amp;quot;Reduced-error pruning&amp;quot; is a fast pruning procedure for decision trees that is known to produce small and accurate trees. Apart from the data from which the tree is grown, it uses an independent &amp;quot;pruning&amp;quot; set, and pruning decisions are based on the model's error rate on this fresh data. Recently it has been observed that reduced-error pruning overfits the pruning data, producing unnecessarily large decision trees. This paper investigates whether standard statistical significance tests can be used to counter this phenomenon.
1132&lt;/p&gt;&lt;/div&gt;
1133
1134
1135
1136&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1137
1138&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1139
1140&lt;/p&gt;&lt;/div&gt;
1141
1142
1143
1144&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1145
1146&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1147The problem of overfitting to the pruning set highlights the need for significance testing. We investigate two classes of test, &amp;quot;parametric&amp;quot; and &amp;quot;non-parametric.&amp;quot; The standard chi-squared statistic can be used both in a parametric test and as the basis for a non-parametric permutation test. In both cases it is necessary to select the significance level at which pruning is applied. We show empirically that both versions of the chi-squared test perform equally well if their significance levels are adjusted appropriately. Using a collection of standard datasets, we show that significance testing improves on standard reduced error pruning if the significance level is tailored to the particular dataset at hand using cross-validation, yielding consistently smaller trees that perform at least as well and sometimes better.
1148&lt;/p&gt;&lt;/div&gt;
1149
1150
1151
1152&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1153
1154&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1155
1156&lt;/p&gt;&lt;/div&gt;
1157
1158
1159
1160&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1161
1162&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
116399/11
1164&lt;/p&gt;&lt;/div&gt;
1165
1166
1167
1168&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1169
1170&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1171Weka: Practical machine learning tools and techniques with Java implementations
1172&lt;/p&gt;&lt;/div&gt;
1173
1174
1175
1176&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1177
1178&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1179Ian H. Witten, Eibe Frank, Len Trigg, Mark Hall, Geoffrey Holmes, Sally Jo Cunningham
1180&lt;/p&gt;&lt;/div&gt;
1181
1182
1183
1184&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1185
1186&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1187The Waikato Environment for Knowledge Analysis (Weka) is a comprehensive suite of Java class libraries that implement many state-of-the-art machine learning and data mining algorithms. Weka is freely available on the World-Wide Web and accompanies a new text on data mining [1] which documents and fully explains all the algorithms it contains. Applications written using the Weka class libraries can be run on any computer with a Web browsing capability; this allows users to apply machine learning techniques to their own data regardless of computer platform.
1188&lt;/p&gt;&lt;/div&gt;
1189
1190
1191
1192&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1193
1194&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1195
1196&lt;/p&gt;&lt;/div&gt;
1197
1198
1199
1200&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1201
1202&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
120399/12
1204&lt;/p&gt;&lt;/div&gt;
1205
1206
1207
1208&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1209
1210&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1211Pace Regression
1212&lt;/p&gt;&lt;/div&gt;
1213
1214
1215
1216&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1217
1218&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1219Yong Wang, Ian H. Witten
1220&lt;/p&gt;&lt;/div&gt;
1221
1222
1223
1224&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1225
1226&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1227This paper articulates a new method of linear regression, &amp;ldquo;pace regression&amp;rdquo;, that addresses many drawbacks of standard regression reported in the literature&amp;mdash;particularly the subset selection problem. Pace regression improves on classical ordinary least squares (OLS) regression by evaluating the effect of each variable and using a clustering analysis to improve the statistical basis for estimating their contribution to the overall regression. As well as outperforming OLS, it also outperforms&amp;mdash;in a remarkably general sense&amp;mdash;other linear modeling techniques in the literature, including subset selection procedures, which seek a reduction in dimensionality that falls out as a natural byproduct of pace regression. The paper defines six procedures that share the fundamental idea of pace regression, all of which are theoretically justified in terms of asymptotic performance. Experiments confirm the performance improvement over other techniques.
1228&lt;/p&gt;&lt;/div&gt;
1229
1230
1231
1232&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1233
1234&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1235
1236&lt;/p&gt;&lt;/div&gt;
1237
1238
1239
1240&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1241
1242&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
124399/13
1244&lt;/p&gt;&lt;/div&gt;
1245
1246
1247
1248&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1249
1250&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1251A compression-based algorithm for Chinese word segmentation
1252&lt;/p&gt;&lt;/div&gt;
1253
1254
1255
1256&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1257
1258&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1259W.J. Teahan, Yingying Wen, Rodger McNab, Ian H. Witten
1260&lt;/p&gt;&lt;/div&gt;
1261
1262
1263
1264&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1265
1266&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1267The Chinese language is written without using spaces or other word delimiters. Although a text may be thought of as a corresponding sequence of words, there is considerable ambiguity in the placement of boundaries. Interpreting a text as a sequence of words is beneficial for some information retrieval and storage tasks: for example, full-text search, word-based compression, and keyphrase extraction.
1268&lt;/p&gt;&lt;/div&gt;
1269
1270
1271
1272&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1273
1274&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1275
1276&lt;/p&gt;&lt;/div&gt;
1277
1278
1279
1280&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1281
1282&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1283We describe a scheme that infers appropriate positions for word boundaries using an adaptive language model that is standard in text compression. It is trained on a corpus of pre-segmented text, and when applied to new text, interpolates word boundaries so as to maximize the compression obtained. This simple and general method performs well with respect to specialized schemes for Chinese language segmentation.
1284&lt;/p&gt;&lt;/div&gt;
1285
1286
1287
1288&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1289
1290&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1291
1292&lt;/p&gt;&lt;/div&gt;
1293
1294
1295
1296&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1297
1298&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
129999/14
1300&lt;/p&gt;&lt;/div&gt;
1301
1302
1303
1304&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1305
1306&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1307Clustering with finite data from semi-parametric mixture distributions
1308&lt;/p&gt;&lt;/div&gt;
1309
1310
1311
1312&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1313
1314&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1315Yong Wang, Ian H. Witten
1316&lt;/p&gt;&lt;/div&gt;
1317
1318
1319
1320&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1321
1322&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1323Existing clustering methods for the semi-parametric mixture distribution perform well as the volume of data increases. However, they all suffer from a serious drawback in finite-data situations: small outlying groups of data points can be completely ignored in the clusters that are produced, no matter how far away they lie from the major clusters. This can result in unbounded loss if the loss function is sensitive to the distance between clusters.
1324&lt;/p&gt;&lt;/div&gt;
1325
1326
1327
1328&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1329
1330&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1331This paper proposes a new distance-based clustering method that overcomes the problem by avoiding global constraints. Experimental results illustrate its superiority to existing methods when small clusters are present in finite data sets; they also suggest that it is more accurate and stable than other methods even when there are no small clusters.
1332&lt;/p&gt;&lt;/div&gt;
1333
1334
1335
1336&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1337
1338&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1339
1340&lt;/p&gt;&lt;/div&gt;
1341
1342
1343
1344&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1345
1346&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
134799/15
1348&lt;/p&gt;&lt;/div&gt;
1349
1350
1351
1352&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1353
1354&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1355
1356&lt;/p&gt;&lt;/div&gt;
1357
1358
1359
1360&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1361
1362&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1363
1364&lt;/p&gt;&lt;/div&gt;
1365
1366
1367
1368&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1369
1370&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
137199/16
1372&lt;/p&gt;&lt;/div&gt;
1373
1374
1375
1376&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1377
1378&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1379The Niupepa Collection: Opening the blinds on a window to the past
1380&lt;/p&gt;&lt;/div&gt;
1381
1382
1383
1384&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1385
1386&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1387Te Taka Keegan, Sally Jo Cunningham, Mark Apperley
1388&lt;/p&gt;&lt;/div&gt;
1389
1390
1391
1392&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1393
1394&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1395This paper describes the building of a digital library collection of historic newspapers. The newspapers (&lt;i&gt;Niupepa&lt;/i&gt; in Maori), which were published in New Zealand during the period 1842 to 1933, form a unique historical record of the Maori language, and of events from an historical perspective. Images of these newspapers have been converted to digital form, electronic text extracted from these, and the collection is now being made available over the Internet as a part of the New Zealand Digital Library (NZDL) project at the University of Waikato.
1396&lt;/p&gt;&lt;/div&gt;
1397
1398
1399
1400&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1401
1402&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1403
1404&lt;/p&gt;&lt;/div&gt;
1405
1406
1407
1408&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1409
1410&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1411
1412&lt;/p&gt;&lt;/div&gt;
1413
1414
1415
1416&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1417
1418&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1419
1420&lt;/p&gt;&lt;/div&gt;
1421
1422
1423
1424&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1425
1426&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
142798/1
1428&lt;/p&gt;&lt;/div&gt;
1429
1430
1431
1432&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1433
1434&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1435Boosting trees for cost-sensitive classifications
1436&lt;/p&gt;&lt;/div&gt;
1437
1438
1439
1440&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1441
1442&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1443Kai Ming Ting, Zijian Zheng
1444&lt;/p&gt;&lt;/div&gt;
1445
1446
1447
1448&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1449
1450&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1451This paper explores two boosting techniques for cost-sensitive tree classification in the situation where misclassification costs change very often. Ideally, one would like to have only one induction, and use the induced model for different misclassification costs. Thus, it demands robustness of the induced model against cost changes. Combining multiple trees gives robust predictions against this change. We demonstrate that ordinary boosting combined with the minimum expected cost criterion to select the prediction class is a good solution under this situation. We also introduce a variant of the ordinary boosting procedure which utilizes the cost information during training. We show that the proposed technique performs better than the ordinary boosting in terms of misclassification cost. However, this technique requires to induce a set of new trees every time the cost changes. Our empirical investigation also reveals some interesting behavior of boosting decision trees for cost-sensitive classification.
1452&lt;/p&gt;&lt;/div&gt;
1453
1454
1455
1456&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1457
1458&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1459
1460&lt;/p&gt;&lt;/div&gt;
1461
1462
1463
1464&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1465
1466&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1467
1468&lt;/p&gt;&lt;/div&gt;
1469
1470
1471
1472&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1473
1474&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
147598/2
1476&lt;/p&gt;&lt;/div&gt;
1477
1478
1479
1480&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1481
1482&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1483Generating accurate rule sets without global optimization
1484&lt;/p&gt;&lt;/div&gt;
1485
1486
1487
1488&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1489
1490&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1491Eibe Frank, Ian H. Witten
1492&lt;/p&gt;&lt;/div&gt;
1493
1494
1495
1496&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1497
1498&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1499The two dominant schemes for rule-learning, C4.5 and RIPPER, both operate in two stages. First they induce an initial rule set and then they refine it using a rather complex optimization stage that discards (C4.5) or adjusts (RIPPER) individual rules to make them work better together. In contrast, this paper shows how good rule sets can be learned one rule at a time, without any need for global optimization. We present an algorithm for inferring rules by repeatedly generating partial decision trees, thus combining the two major paradigms for rule generation-creating rules from decision trees and the separate-and-conquer rule-learning technique. The algorithm is straightforward and elegant: despite this, experiments on standard datasets show that it produces rule sets that are as accurate as and of similar size to those generated by C4.5, and more accurate than RIPPER's. Moreover, it operates efficiently, and because it avoids postprocessing, does not suffer the extremely slow performance on pathological example sets for which the C4.5 method has been criticized.
1500&lt;/p&gt;&lt;/div&gt;
1501
1502
1503
1504&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1505
1506&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1507
1508&lt;/p&gt;&lt;/div&gt;
1509
1510
1511
1512&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1513
1514&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1515
1516&lt;/p&gt;&lt;/div&gt;
1517
1518
1519
1520&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1521
1522&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
152398/3
1524&lt;/p&gt;&lt;/div&gt;
1525
1526
1527
1528&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1529
1530&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1531VQuery: a graphical user interface for Boolean query Specification and dynamic result preview
1532&lt;/p&gt;&lt;/div&gt;
1533
1534
1535
1536&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1537
1538&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1539Steve Jones
1540&lt;/p&gt;&lt;/div&gt;
1541
1542
1543
1544&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1545
1546&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1547Textual query languages based on Boolean logic are common amongst the search facilities of on-line information repositories. However, there is evidence to suggest that the syntactic and semantic demands of such languages lead to user errors and adversely affect the time that it takes users to form queries. Additionally, users are faced with user interfaces to these repositories which are unresponsive and uninformative, and consequently fail to support effective query refinement. We suggest that graphical query languages, particularly Venn-like diagrams, provide a natural medium for Boolean query specification which overcomes the problems of textual query languages. Also, dynamic result previews can be seamlessly integrated with graphical query specification to increase the effectiveness of query refinements. We describe VQuery, a query interface to the New Zealand Digital Library which exploits querying by Venn diagrams and integrated query result previews.
1548&lt;/p&gt;&lt;/div&gt;
1549
1550
1551
1552&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1553
1554&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1555
1556&lt;/p&gt;&lt;/div&gt;
1557
1558
1559
1560&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1561
1562&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1563
1564&lt;/p&gt;&lt;/div&gt;
1565
1566
1567
1568&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1569
1570&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
157198/4
1572&lt;/p&gt;&lt;/div&gt;
1573
1574
1575
1576&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1577
1578&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1579Revising &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt;: semantics and logic
1580&lt;/p&gt;&lt;/div&gt;
1581
1582
1583
1584&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1585
1586&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1587Martin C. Henson, Steve Reeves
1588&lt;/p&gt;&lt;/div&gt;
1589
1590
1591
1592&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1593
1594&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1595We introduce a simple specification logic &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt;c comprising a logic and semantics (in &amp;lt;I&amp;gt;ZF&amp;lt;/I&amp;gt; set theory). We then provide an interpretation for (a rational reconstruction of) the specification language &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt; within &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt;c. As a result we obtain a sound logic for &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt;, including the schema calculus. A consequence of our formalisation is a critique of a number of concepts used in &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt;. We demonstrate that the complications and confusions which these concepts introduce can be avoided without compromising expressibility.
1596&lt;/p&gt;&lt;/div&gt;
1597
1598
1599
1600&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1601
1602&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1603
1604&lt;/p&gt;&lt;/div&gt;
1605
1606
1607
1608&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1609
1610&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1611
1612&lt;/p&gt;&lt;/div&gt;
1613
1614
1615
1616&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1617
1618&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
161998/5
1620&lt;/p&gt;&lt;/div&gt;
1621
1622
1623
1624&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1625
1626&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1627A logic for the schema calculus
1628&lt;/p&gt;&lt;/div&gt;
1629
1630
1631
1632&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1633
1634&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1635Martin C. Henson, Steve Reeves
1636&lt;/p&gt;&lt;/div&gt;
1637
1638
1639
1640&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1641
1642&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1643In this paper we introduce and investigate a logic for the schema calculus of &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt;. The schema calculus is arguably the reason for &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt;'s popularity but so far no true calculus (a sound system of rules for reasoning about schema expressions) has been given. Presentations thus far have either failed to provide a calculus (e.g. the draft standard [3]) or have fallen back on informal descriptions at a syntactic level (most text books e.g. [7[). Once the calculus is established we introduce a derived equational logic which enables us to formalise properly the informal notations of schema expression equality to be found in the literature.
1644&lt;/p&gt;&lt;/div&gt;
1645
1646
1647
1648&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1649
1650&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1651
1652&lt;/p&gt;&lt;/div&gt;
1653
1654
1655
1656&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1657
1658&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1659
1660&lt;/p&gt;&lt;/div&gt;
1661
1662
1663
1664&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1665
1666&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
166798/6
1668&lt;/p&gt;&lt;/div&gt;
1669
1670
1671
1672&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1673
1674&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1675New foundations for &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt;
1676&lt;/p&gt;&lt;/div&gt;
1677
1678
1679
1680&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1681
1682&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1683Martin C. Henson, Steve Reeves
1684&lt;/p&gt;&lt;/div&gt;
1685
1686
1687
1688&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1689
1690&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1691We provide a constructive and intensional interpretation for the specification language &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt; in a theory of operations and kinds &amp;lt;I&amp;gt;T&amp;lt;/I&amp;gt;. The motivation is to facilitate the development of an integrated approach to program construction. We illustrate the new foundations for &amp;lt;I&amp;gt;Z&amp;lt;/I&amp;gt; with examples.
1692&lt;/p&gt;&lt;/div&gt;
1693
1694
1695
1696&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1697
1698&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1699
1700&lt;/p&gt;&lt;/div&gt;
1701
1702
1703
1704&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1705
1706&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1707
1708&lt;/p&gt;&lt;/div&gt;
1709
1710
1711
1712&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1713
1714&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
171598/7
1716&lt;/p&gt;&lt;/div&gt;
1717
1718
1719
1720&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1721
1722&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1723Predicting apple bruising relationships using machine learning
1724&lt;/p&gt;&lt;/div&gt;
1725
1726
1727
1728&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1729
1730&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1731G. Holmes, S.J. Cunningham, B.T. Dela Rue, A.F. Bollen
1732&lt;/p&gt;&lt;/div&gt;
1733
1734
1735
1736&lt;p&gt;&lt;div name=&quot;Body Text&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1737
1738&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1739Many models have been used to describe the influence of internal or external factors on apple bruising. Few of these have addressed the application of derived relationships to the evaluation of commercial operations. From an industry perspective, a model must enable fruit to be rejected on the basis of a commercially significant bruise and must also accurately quantify the effects of various combinations of input features (such as cultivar, maturity, size, and so on) on bruise prediction. Input features must in turn have characteristics which are measurable commercially; for example, the measure of force should be impact energy rather than energy absorbed. Further, as the commercial criteria for acceptable damage levels change, the model should be versatile enough to regenerate new bruise thresholds from existing data.
1740&lt;/p&gt;&lt;/div&gt;
1741
1742
1743
1744&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1745
1746&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1747
1748&lt;/p&gt;&lt;/div&gt;
1749
1750
1751
1752&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1753
1754&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1755Machine learning is a burgeoning technology with a vast range of potential applications particularly in agriculture where large amounts of data can be readily collected [1]. The main advantage of using a machine learning method in an application is that the models built for prediction can be viewed and understood by the owner of the data who is in a position to determine the usefulness of the model, an essential component in a commercial environment.
1756&lt;/p&gt;&lt;/div&gt;
1757
1758
1759
1760&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1761
1762&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1763
1764&lt;/p&gt;&lt;/div&gt;
1765
1766
1767
1768&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1769
1770&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1771
1772&lt;/p&gt;&lt;/div&gt;
1773
1774
1775
1776&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1777
1778&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
177998/8
1780&lt;/p&gt;&lt;/div&gt;
1781
1782
1783
1784&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1785
1786&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1787An evaluation of passage-level indexing strategies for a technical report archive
1788&lt;/p&gt;&lt;/div&gt;
1789
1790
1791
1792&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1793
1794&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1795Michael Williams
1796&lt;/p&gt;&lt;/div&gt;
1797
1798
1799
1800&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1801
1802&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1803Past research has shown that using evidence from document passages rather than complete documents is an effective way of improving the precision of full-text database searches. However, passage-level indexing has yet to be widely adopted for commercial or online databases.
1804&lt;/p&gt;&lt;/div&gt;
1805
1806
1807
1808&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1809
1810&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1811
1812&lt;/p&gt;&lt;/div&gt;
1813
1814
1815
1816&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1817
1818&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1819This paper reports on experiments designed to test the efficacy of passage-level indexing with a particular collection of a full-text online database, the New Zealand Digital Library. Discourse passages and word-window passages are used for the indexing process. Both ranked and Boolean searching are used to test the resulting indexes.
1820&lt;/p&gt;&lt;/div&gt;
1821
1822
1823
1824&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1825
1826&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1827
1828&lt;/p&gt;&lt;/div&gt;
1829
1830
1831
1832&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1833
1834&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1835Overlapping window passages are shown to offer the best retrieval performance with both ranked and Boolean queries. Modifications may be necessary to the term weighting methodology in order to ensure optimal ranked query performance.
1836&lt;/p&gt;&lt;/div&gt;
1837
1838
1839
1840&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1841
1842&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1843
1844&lt;/p&gt;&lt;/div&gt;
1845
1846
1847
1848&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1849
1850&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1851
1852&lt;/p&gt;&lt;/div&gt;
1853
1854
1855
1856&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1857
1858&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
185998/9
1860&lt;/p&gt;&lt;/div&gt;
1861
1862
1863
1864&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1865
1866&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1867Managing multiple collections, multiple languages, and multiple media in a distributed digital library
1868&lt;/p&gt;&lt;/div&gt;
1869
1870
1871
1872&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1873
1874&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1875Ian H. Witten, Rodger McNab, Steve Jones, Sally Jo Cunningham, David Bainbridge, Mark Apperley
1876&lt;/p&gt;&lt;/div&gt;
1877
1878
1879
1880&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1881
1882&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1883Managing the organizational and software complexity of a comprehensive digital library presents a significant challenge. Different library collections each have their own distinctive features. Different presentation languages have structural implications such as left-to-right writing order and text-only interfaces for the visually impaired. Different media involve different file formats, and-more importantly-radically different search strategies are required for non-textual media. In a distributed library, new collections can appear asynchronously on servers in different parts of the world. And as searching interfaces mature from the command-line era exemplified by current Web search engines into the age of reactive visual interfaces, experimental new interfaces must be developed, supported, and tested. This paper describes our experience, gained from operating a substantial digital library service over several years, in solving these problems by designing an appropriate software architecture.
1884&lt;/p&gt;&lt;/div&gt;
1885
1886
1887
1888&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1889
1890&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1891
1892&lt;/p&gt;&lt;/div&gt;
1893
1894
1895
1896&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1897
1898&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1899
1900&lt;/p&gt;&lt;/div&gt;
1901
1902
1903
1904&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1905
1906&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
190798/10
1908&lt;/p&gt;&lt;/div&gt;
1909
1910
1911
1912&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1913
1914&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1915Experiences with a weighted decision tree learner
1916&lt;/p&gt;&lt;/div&gt;
1917
1918
1919
1920&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1921
1922&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1923John G. Cleary, Leonard E. Trigg
1924&lt;/p&gt;&lt;/div&gt;
1925
1926
1927
1928&lt;p&gt;&lt;div name=&quot;Body Text&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1929
1930&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1931Machine learning algorithms for inferring decision trees typically choose a single &amp;ldquo;best&amp;rdquo; tree to describe the training data. Recent research has shown that classification performance can be significantly improved by voting predictions of multiple, independently produced decision trees. This paper describes an algorithm, OB1, that makes a weighted sum over many possible models. We describe one instance of OB1, that includes &amp;lt;I&amp;gt;all&amp;lt;/I&amp;gt; possible decision trees as well as naïve Bayesian models. OB1 is compared with a number of other decision tree and instance based learning alogrithms on some of the data sets from the UCI repository. Both an information gain and an accuracy measure are used for the comparison. On the information gain measure OB1 performs significantly better than all the other algorithms. On the accuracy measure it is significantly better than all the algorithms except naïve Bayes which performs comparably to OB1.
1932&lt;/p&gt;&lt;/div&gt;
1933
1934
1935
1936&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1937
1938&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1939
1940&lt;/p&gt;&lt;/div&gt;
1941
1942
1943
1944&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1945
1946&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1947
1948&lt;/p&gt;&lt;/div&gt;
1949
1950
1951
1952&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1953
1954&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
195598/11
1956&lt;/p&gt;&lt;/div&gt;
1957
1958
1959
1960&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1961
1962&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1963An entropy gain measure of numeric prediction performance
1964&lt;/p&gt;&lt;/div&gt;
1965
1966
1967
1968&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1969
1970&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1971Leonard Trigg
1972&lt;/p&gt;&lt;/div&gt;
1973
1974
1975
1976&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1977
1978&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1979Categorical classifier performance is typically evaluated with respect to error rate, expressed as a percentage of test instances that were not correctly classified. When a classifier produces multiple classifications for a test instance, the prediction is counted as incorrect (even if the correct class was one of the predictions). Although commonly used in the literature, error rate is a coarse measure of classifier performance, as it is based only on a single prediction offered for a test instance. Since many classifiers can produce a class distribution as a prediction, we should use this to provide a better measure of how much information the classifier is extracting from the domain.
1980&lt;/p&gt;&lt;/div&gt;
1981
1982
1983
1984&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1985
1986&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1987
1988&lt;/p&gt;&lt;/div&gt;
1989
1990
1991
1992&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
1993
1994&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
1995Numeric classifiers are a relatively new development in machine learning, and as such there is no single performance measure that has become standard. Typically these machine learning schemes predict a single real number for each test instance, and the error between the predicted and actual value is used to calculate a myriad of performance measures such as correlation coefficient, root mean squared error, mean absolute error, relative absolute error, and root relative squared error. With so many performance measures it is difficult to establish an overall performance evaluation.
1996&lt;/p&gt;&lt;/div&gt;
1997
1998
1999
2000&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2001
2002&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2003
2004&lt;/p&gt;&lt;/div&gt;
2005
2006
2007
2008&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2009
2010&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2011The next section describes a performance measure for machine learning schemes that attempts to overcome the problems with current measures. In addition, the same evaluation measure is used for categorical and numeric classifier.
2012&lt;/p&gt;&lt;/div&gt;
2013
2014
2015
2016&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2017
2018&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2019
2020&lt;/p&gt;&lt;/div&gt;
2021
2022
2023
2024&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2025
2026&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2027
2028&lt;/p&gt;&lt;/div&gt;
2029
2030
2031
2032&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2033
2034&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2035
2036&lt;/p&gt;&lt;/div&gt;
2037
2038
2039
2040&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2041
2042&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
204398/12
2044&lt;/p&gt;&lt;/div&gt;
2045
2046
2047
2048&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2049
2050&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2051Proceedings of CBISE '98 CaiSE*98 Workshop on Component Based Information Systems Engineering
2052&lt;/p&gt;&lt;/div&gt;
2053
2054
2055
2056&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2057
2058&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2059Edited by John Grundy
2060&lt;/p&gt;&lt;/div&gt;
2061
2062
2063
2064&lt;p&gt;&lt;div name=&quot;Body Text&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2065
2066&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2067Component-based information systems development is an area of research and practice of increasing importance. Information Systems developers have realised that traditional approaches to IS engineering produce monolithic, difficult to maintain, difficult to reuse systems. In contrast, the use of software components, which embody data, functionality and well-specified and understood interfaces, makes interoperable, distributed and highly reusable IS components feasible. Component-based approaches to IS engineering can be used at strategic and organisational levels, to model business processes and whole IS architectures, in development methods which utilise component-based models during analysis and design, and in system implementation. Reusable components can allow end users to compose and configure their own Information Systems, possibly from a range of suppliers, and to more tightly couple their organisational workflows with their IS support.
2068&lt;/p&gt;&lt;/div&gt;
2069
2070
2071
2072&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2073
2074&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2075
2076&lt;/p&gt;&lt;/div&gt;
2077
2078
2079
2080&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2081
2082&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2083This workshop proceedings contains a range of papers addressing one or more of the above issues relating to the use of component models for IS development. All of these papers were refereed by at least two members of an international workshop committee comprising industry and academic researchers and users of component technologies. Strategic uses of components are addressed in the first three papers, while the following three address uses of components for systems design and workflow management. Systems development using components, and the provision of environments for component management are addressed in the following group of five papers. The last three papers in this proceedings address component management and analysis techniques.
2084&lt;/p&gt;&lt;/div&gt;
2085
2086
2087
2088&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2089
2090&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2091
2092&lt;/p&gt;&lt;/div&gt;
2093
2094
2095
2096&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2097
2098&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2099All of these papers provide new insights into the many varied uses of component technology for IS engineering. I hope you find them as interesting and useful as I have when collating this proceedings and organising the workshop.
2100&lt;/p&gt;&lt;/div&gt;
2101
2102
2103
2104&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2105
2106&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2107
2108&lt;/p&gt;&lt;/div&gt;
2109
2110
2111
2112&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2113
2114&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2115
2116&lt;/p&gt;&lt;/div&gt;
2117
2118
2119
2120&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2121
2122&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
212398/13
2124&lt;/p&gt;&lt;/div&gt;
2125
2126
2127
2128&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2129
2130&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2131An analysis of usage of a digital library
2132&lt;/p&gt;&lt;/div&gt;
2133
2134
2135
2136&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2137
2138&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2139Steve Jones, Sally Jo Cunningham, Rodger McNab
2140&lt;/p&gt;&lt;/div&gt;
2141
2142
2143
2144&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2145
2146&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2147As experimental digital library testbeds gain wider acceptance and develop significant user bases, it becomes important to investigate the ways in which users interact with the systems in practice. Transaction logs are one source of usage information, and the information on user behaviour can be culled from them both automatically (through calculation of summary statistics) and manually (by examining query strings for semantic clues on search motivations and searching strategy). We conduct a transaction log analysis on user activity in the Computer Science Technical Reports Collection of the New Zealand Digital Library, and report insights gained and identify resulting search interface design issues.
2148&lt;/p&gt;&lt;/div&gt;
2149
2150
2151
2152&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2153
2154&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2155
2156&lt;/p&gt;&lt;/div&gt;
2157
2158
2159
2160&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2161
2162&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2163
2164&lt;/p&gt;&lt;/div&gt;
2165
2166
2167
2168&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2169
2170&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
217198/14
2172&lt;/p&gt;&lt;/div&gt;
2173
2174
2175
2176&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2177
2178&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2179Measuring ATM traffic: final report for New Zealand Telecom
2180&lt;/p&gt;&lt;/div&gt;
2181
2182
2183
2184&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2185
2186&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2187John Cleary, Ian Graham, Murray Pearson, Tony McGregor
2188&lt;/p&gt;&lt;/div&gt;
2189
2190
2191
2192&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2193
2194&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2195The report describes the development of a low-cost ATM monitoring system, hosted by a standard PC. The monitor can be used remotely returning information on ATM traffic flows to a central site. The monitor is interfaces to a GPS timing receiver, which provides an absolute time accuracy of better than 1 usec. By monitoring the same traffic flow at different points in a network it is possible to measure cell delay and delay variation in real time, and with existing traffic. The monitoring system characterises cells by a CRC calculated over the cell payload, thus special measurement cells are not required. Delays in both local area and wide-area networks have been measured using this system. It is possible to measure delay in a network that is not end-to-end ATM, as long as some cells remain identical at the entry and exit points. Examples are given of traffic and delay measurements in both wide and local area network systems, including delays measured over the Internet from Canada to New Zealand.
2196&lt;/p&gt;&lt;/div&gt;
2197
2198
2199
2200&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2201
2202&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2203
2204&lt;/p&gt;&lt;/div&gt;
2205
2206
2207
2208&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2209
2210&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2211
2212&lt;/p&gt;&lt;/div&gt;
2213
2214
2215
2216&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2217
2218&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2219
2220&lt;/p&gt;&lt;/div&gt;
2221
2222
2223
2224&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2225
2226&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
222798/15
2228&lt;/p&gt;&lt;/div&gt;
2229
2230
2231
2232&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2233
2234&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2235Despite its simplicity, the naïve Bayes learning scheme performs well on most classification tasks, and is often significantly more accurate than more sophisticated methods. Although the probability estimates that it produces can be inaccurate, it often assigns maximum probability to the correct class. This suggests that its good performance might be restricted to situations where the output is categorical. It is therefore interesting to see how it performs in domains where the predicted value is numeric, because in this case, predictions are more sensitive to inaccurate probability estimates.&amp;lt;P&amp;gt;
2236&lt;/p&gt;&lt;/div&gt;
2237
2238
2239
2240&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2241
2242&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2243
2244&lt;/p&gt;&lt;/div&gt;
2245
2246
2247
2248&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2249
2250&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2251This paper shows how to apply the naïve Bayes methodology to numeric prediction (i.e. regression) tasks, and compares it to linear regression, instance-based learning, and a method that produces &amp;ldquo;model trees&amp;rdquo;-decision trees with linear regression functions at the leaves. Although we exhibit an artificial dataset for which naïve Bayes is the method of choice, on real-world datasets it is almost uniformly worse than model trees. The comparison with linear regression depends on the error measure: for one measure naïve Bayes performs similarly, for another it is worse. Compared to instance-based learning, it performs similarly with respect to both measures. These results indicate that the simplistic statistical assumption that naïve Bayes makes is indeed more restrictive for regression than for classification.
2252&lt;/p&gt;&lt;/div&gt;
2253
2254
2255
2256&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2257
2258&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2259
2260&lt;/p&gt;&lt;/div&gt;
2261
2262
2263
2264&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2265
2266&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
226798/16
2268&lt;/p&gt;&lt;/div&gt;
2269
2270
2271
2272&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2273
2274&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2275Link as you type: using key phrases for automated dynamic link generation
2276&lt;/p&gt;&lt;/div&gt;
2277
2278
2279
2280&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2281
2282&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2283Steve Jones
2284&lt;/p&gt;&lt;/div&gt;
2285
2286
2287
2288&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2289
2290&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2291When documents are collected together from diverse sources they are unlikely to contain useful hypertext links to support browsing amongst them. For large collections of thousands of documents it is prohibitively resource intensive to manually insert links into each document. Users of such collections may wish to relate documents within them to text that they are themselves generating. This process, often involving keyword searching, distracts from the authoring process and results in material related to query terms but not necessarily to the author's document. Query terms that are effective in one collection might not be so in another. We have developed Phrasier, a system that integrates authoring (of text and hyperlinks), browsing, querying and reading in support of information retrieval activities. Phrasier exploits key phrases which are automatically extracted from documents in a collection, and uses them as link anchors and to identify candidate destinations for hyperlinks. This system suggests links into existing collections for purposes of authoring and retrieval of related information, creates links between documents in a collection and provides supportive document and link overviews.
2292&lt;/p&gt;&lt;/div&gt;
2293
2294
2295
2296&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2297
2298&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2299
2300&lt;/p&gt;&lt;/div&gt;
2301
2302
2303
2304&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2305
2306&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2307
2308&lt;/p&gt;&lt;/div&gt;
2309
2310
2311
2312&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2313
2314&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
231598/17
2316&lt;/p&gt;&lt;/div&gt;
2317
2318
2319
2320&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2321
2322&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2323Melody based tune retrieval over the World Wide Web
2324&lt;/p&gt;&lt;/div&gt;
2325
2326
2327
2328&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2329
2330&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2331David Bainbridge, Rodger J. McNab, Lloyd A. Smith
2332&lt;/p&gt;&lt;/div&gt;
2333
2334
2335
2336&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2337
2338&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2339In this paper we describe the steps taken to develop a Web-based version of an existing stand-alone, single-user digital library application for melodical searching of a collection of music. For the three key components: input, searching, and output, we assess the suitability of various Web-based strategies that deal with the now distributed software architecture and explain the decisions we made. The resulting melody indexing service, known as MELDEX, has been in operation for one year, and the feed-back we have received has been favorable.
2340&lt;/p&gt;&lt;/div&gt;
2341
2342
2343
2344&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2345
2346&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2347
2348&lt;/p&gt;&lt;/div&gt;
2349
2350
2351
2352&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2353
2354&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
235598/18
2356&lt;/p&gt;&lt;/div&gt;
2357
2358
2359
2360&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2361
2362&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2363Making oral history accessible over the World Wide Web
2364&lt;/p&gt;&lt;/div&gt;
2365
2366
2367
2368&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2369
2370&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2371David Bainbridge, Sally Jo Cunningham
2372&lt;/p&gt;&lt;/div&gt;
2373
2374
2375
2376&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2377
2378&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2379We describe a multimedia, WWW-based oral history collection constructed from off-the-shelf or publicly available software. The source materials for the collection include audio tapes of interviews and summary transcripts of each interview, as well as photographs illustrating episodes mentioned in the tapes. Sections of the transcripts are manually matched to associated segments of the tapes, and the tapes are digitized. Users search a full-text retrieval system based on the text transcripts to retrieve relevant transcript sections and their associated audio recordings and photographs. It is also possible to search for photos by matching text queries against text descriptions of the photos in the collection, where the located photos link back to their respective interview transcript and audio recordings.
2380&lt;/p&gt;&lt;/div&gt;
2381
2382
2383
2384&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2385
2386&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2387
2388&lt;/p&gt;&lt;/div&gt;
2389
2390
2391
2392&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2393
2394&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2395
2396&lt;/p&gt;&lt;/div&gt;
2397
2398
2399
2400&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2401
2402&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2403
2404&lt;/p&gt;&lt;/div&gt;
2405
2406
2407
2408&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2409
2410&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2411
2412&lt;/p&gt;&lt;/div&gt;
2413
2414
2415
2416&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2417
2418&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2419&lt;b&gt;1997&lt;/b&gt;
2420&lt;/p&gt;&lt;/div&gt;
2421
2422
2423
2424&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2425
2426&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2427
2428&lt;/p&gt;&lt;/div&gt;
2429
2430
2431
2432&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2433
2434&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
243597/1
2436&lt;/p&gt;&lt;/div&gt;
2437
2438
2439
2440&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2441
2442&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2443A dynamic and flexible representation of social relationships in CSCW
2444&lt;/p&gt;&lt;/div&gt;
2445
2446
2447
2448&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2449
2450&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2451Steve Jones, Steve Marsh
2452&lt;/p&gt;&lt;/div&gt;
2453
2454
2455
2456&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2457
2458&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2459CSCW system designers lack effective support in addressing the social issues and interpersonal relationships which are linked with the use of CSCW systems. We present a formal description of trust to support CSCW system designers in considering the social aspects of group work, embedding those considerations in systems and analysing computer supported group processes.
2460&lt;/p&gt;&lt;/div&gt;
2461
2462
2463
2464&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2465
2466&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2467
2468&lt;/p&gt;&lt;/div&gt;
2469
2470
2471
2472&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2473
2474&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2475We argue that trust is a critical aspect in group work, and describe what we consider to be the building blocks of trust. We then present a formal notation for the building blocks, their use in reasoning about social interactions and how they are amended over time.
2476&lt;/p&gt;&lt;/div&gt;
2477
2478
2479
2480&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2481
2482&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2483
2484&lt;/p&gt;&lt;/div&gt;
2485
2486
2487
2488&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2489
2490&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2491We then consider how the formalism may be used in practice, and present some insights from initial analysis of the behaviour of the formalism. This is followed by a description of possible amendments and extensions to the formalism. We conclude that it is possible to formalise a notion of trust and to model the formalisation by a computational mechanism.
2492&lt;/p&gt;&lt;/div&gt;
2493
2494
2495
2496&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2497
2498&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2499
2500&lt;/p&gt;&lt;/div&gt;
2501
2502
2503
2504&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2505
2506&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2507
2508&lt;/p&gt;&lt;/div&gt;
2509
2510
2511
2512&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2513
2514&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
251597/2
2516&lt;/p&gt;&lt;/div&gt;
2517
2518
2519
2520&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2521
2522&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2523Design issues for World Wide Web navigation visualisation tools
2524&lt;/p&gt;&lt;/div&gt;
2525
2526
2527
2528&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2529
2530&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2531Andy Cockburn, Steve Jones
2532&lt;/p&gt;&lt;/div&gt;
2533
2534
2535
2536&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2537
2538&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2539The World Wide Web (WWW) is a successful hypermedia information space used by millions of people, yet it suffers from many deficiencies and problems in support for navigation around its vast information space. In this paper we identify the origins of these navigation problems, namely WWW browser design, WWW page design, and WWW page description languages. Regardless of their origins, these problems are eventually represented to the user at the browser's user interface. To help overcome these problems, many tools are being developed which allow users to visualise WWW subspaces. We identify five key issues in the design and functionality of these visualisation systems: characteristics of the visual representation, the scope of the subspace representation, the mechanisms for generating the visualisation, the degree of browser independence, and the navigation support facilities. We provide a critical review of the diverse range of WWW visualisation tools with respect to these issues.
2540&lt;/p&gt;&lt;/div&gt;
2541
2542
2543
2544&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2545
2546&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2547
2548&lt;/p&gt;&lt;/div&gt;
2549
2550
2551
2552&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2553
2554&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2555
2556&lt;/p&gt;&lt;/div&gt;
2557
2558
2559
2560&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2561
2562&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
256397/3
2564&lt;/p&gt;&lt;/div&gt;
2565
2566
2567
2568&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2569
2570&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2571Stacked generalization: when does it work?
2572&lt;/p&gt;&lt;/div&gt;
2573
2574
2575
2576&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2577
2578&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2579Kai Ming Ting, Ian H. Witten
2580&lt;/p&gt;&lt;/div&gt;
2581
2582
2583
2584&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2585
2586&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2587Stacked generalization is a general method of using a high-level model to combine lower-level models to achieve greater predictive accuracy. In this paper we address two crucial issues which have been considered to be a 'black art' in classification tasks ever since the introduction of stacked generalization in 1992 by Wolpert: the type of generalizer that is suitable to derive the higher-level model, and the kind of attributes that should be used as its input.
2588&lt;/p&gt;&lt;/div&gt;
2589
2590
2591
2592&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2593
2594&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2595
2596&lt;/p&gt;&lt;/div&gt;
2597
2598
2599
2600&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2601
2602&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2603We demonstrate the effectiveness of stacked generalization for combining three different types of learning algorithms, and also for combining models of the same type derived from a single learning algorithm in a multiple-data-batches scenario. We also compare the performance of stacked generalization with published results arcing and bagging.
2604&lt;/p&gt;&lt;/div&gt;
2605
2606
2607
2608&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2609
2610&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2611
2612&lt;/p&gt;&lt;/div&gt;
2613
2614
2615
2616&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2617
2618&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2619
2620&lt;/p&gt;&lt;/div&gt;
2621
2622
2623
2624&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2625
2626&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
262797/4
2628&lt;/p&gt;&lt;/div&gt;
2629
2630
2631
2632&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2633
2634&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2635Browsing in digital libraries: a phrase-based approach
2636&lt;/p&gt;&lt;/div&gt;
2637
2638
2639
2640&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2641
2642&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2643Craig Nevill-Manning, Ian H. Witten, Gordon W. Paynter
2644&lt;/p&gt;&lt;/div&gt;
2645
2646
2647
2648&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2649
2650&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2651A key question for digital libraries is this: how should one go about becoming familiar with a digital collection, as opposed to a physical one? Digital collections generally present an appearance which is extremely opaque-a screen, typically a Web page, with no indication of what, or how much, lies beyond: whether a carefully-selected collection or a morass of worthless ephemera; whether half a dozen documents or many millions. At least physical collections occupy physical space, present a physical appearance, and exhibit tangible physical organization. When standing on the threshold of a large library one gains a sense of presence and permanence that reflects the care taken in building and maintaining the collection inside. No-one could confuse it with a dung-heap! Yet in the digital world the difference is not so palpable.
2652&lt;/p&gt;&lt;/div&gt;
2653
2654
2655
2656&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2657
2658&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2659
2660&lt;/p&gt;&lt;/div&gt;
2661
2662
2663
2664&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2665
2666&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2667
2668&lt;/p&gt;&lt;/div&gt;
2669
2670
2671
2672&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2673
2674&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2675
2676&lt;/p&gt;&lt;/div&gt;
2677
2678
2679
2680&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2681
2682&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
268397/5
2684&lt;/p&gt;&lt;/div&gt;
2685
2686
2687
2688&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2689
2690&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2691A graphical notation for the design of information visualisations
2692&lt;/p&gt;&lt;/div&gt;
2693
2694
2695
2696&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2697
2698&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2699Matthew C. Humphrey
2700&lt;/p&gt;&lt;/div&gt;
2701
2702
2703
2704&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2705
2706&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2707Visualisations are coherent, graphical expressions of complex information that enhance people's ability to communicate and reason about that information. Yet despite the importance of visualisations in helping people to understand and solve a wide variety of problems, there is a dearth of formal tools and methods for discussing, describing and designing them. Although simple visualisations, such as bar charts and scatterplots, are easily produced by modern interactive software, novel visualisations of multivariate, multirelational data must be expressed in a programming language. The Relational Visualisation Notation is a new, graphical language for designing such highly expressive visualisations that does not use programming constructs. Instead, the notation is based on relational algebra, which is widely used in database query languages, and it is supported by a suite of direct manipulation tools. This article presents the notation and examines the designs of some interesting visualisations.
2708&lt;/p&gt;&lt;/div&gt;
2709
2710
2711
2712&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2713
2714&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2715
2716&lt;/p&gt;&lt;/div&gt;
2717
2718
2719
2720&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2721
2722&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2723
2724&lt;/p&gt;&lt;/div&gt;
2725
2726
2727
2728&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2729
2730&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2731
2732&lt;/p&gt;&lt;/div&gt;
2733
2734
2735
2736&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2737
2738&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
273997/6
2740&lt;/p&gt;&lt;/div&gt;
2741
2742
2743
2744&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2745
2746&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2747Applications of machine learning in information retrieval
2748&lt;/p&gt;&lt;/div&gt;
2749
2750
2751
2752&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2753
2754&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2755Sally Jo Cunningham, James Littin, Ian H. Witten
2756&lt;/p&gt;&lt;/div&gt;
2757
2758
2759
2760&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2761
2762&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2763Information retrieval systems provide access to collections of thousands, or millions, of documents, from which, by providing an appropriate description, users can recover any one. Typically, users iteratively refine the descriptions they provide to satisfy their needs, and retrieval systems can utilize user feedback on selected documents to indicate the accuracy of the description at any stage. The style of description required from the user, and the way it is employed to search the document database, are consequences of the indexing method used for the collection. The index may take different forms, from storing keywords with links to individual documents, to clustering documents under related topics.
2764&lt;/p&gt;&lt;/div&gt;
2765
2766
2767
2768&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2769
2770&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2771
2772&lt;/p&gt;&lt;/div&gt;
2773
2774
2775
2776&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2777
2778&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2779
2780&lt;/p&gt;&lt;/div&gt;
2781
2782
2783
2784&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2785
2786&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2787
2788&lt;/p&gt;&lt;/div&gt;
2789
2790
2791
2792&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2793
2794&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
279597/7
2796&lt;/p&gt;&lt;/div&gt;
2797
2798
2799
2800&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2801
2802&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2803Computer concepts without computers: a first course in computer science
2804&lt;/p&gt;&lt;/div&gt;
2805
2806
2807
2808&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2809
2810&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2811Geoffrey Holmes, Tony C. Smith, William J. Rogers
2812&lt;/p&gt;&lt;/div&gt;
2813
2814
2815
2816&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2817
2818&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2819While some institutions seek to make CS1 curricula more enjoyable by incorporating specialised educational software [1] or by setting more enjoyable programming assignments [2], we have joined the growing number of Computer Science departments that seek to improve the quality of the CS1 experience by focusing student attention away from the computer monitor [3,4]. Sophisticated computing concepts usually reserved for senior level courses are presented in a &amp;lt;I&amp;gt;popular science&amp;lt;/I&amp;gt; manner, and given equal time alongside the essential introductory programming material. By exposing students to a broad range of specific computational problems we endeavour to make the introductory course more interesting and enjoyable, and instil in students a sense of vision for areas they might specialise in as computing majors.
2820&lt;/p&gt;&lt;/div&gt;
2821
2822
2823
2824&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2825
2826&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2827
2828&lt;/p&gt;&lt;/div&gt;
2829
2830
2831
2832&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2833
2834&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2835
2836&lt;/p&gt;&lt;/div&gt;
2837
2838
2839
2840&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2841
2842&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2843
2844&lt;/p&gt;&lt;/div&gt;
2845
2846
2847
2848&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2849
2850&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
285197/8
2852&lt;/p&gt;&lt;/div&gt;
2853
2854
2855
2856&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2857
2858&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2859A sight-singing tutor
2860&lt;/p&gt;&lt;/div&gt;
2861
2862
2863
2864&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2865
2866&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2867Lloyd A. Smith, Rodger J. McNab
2868&lt;/p&gt;&lt;/div&gt;
2869
2870
2871
2872&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2873
2874&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2875This paper describes a computer program designed to aid its users in learning to sight-sing. Sight-singing-the ability to sing music from a score without prior study-is an important skill for musicians and holds a central place in most university music curricula. Its importance to vocalists is obvious; it is also an important skill for instrumentalists and conductors because it develops the aural imagination necessary to judge how the music should sound, when played (Benward and Carr 1991). Furthermore, it is an important skill for amateur musicians, who can save a great deal of rehearsal time through an ability to sing music at sight.
2876&lt;/p&gt;&lt;/div&gt;
2877
2878
2879
2880&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2881
2882&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2883
2884&lt;/p&gt;&lt;/div&gt;
2885
2886
2887
2888&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2889
2890&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2891
2892&lt;/p&gt;&lt;/div&gt;
2893
2894
2895
2896&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2897
2898&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
289997/9
2900&lt;/p&gt;&lt;/div&gt;
2901
2902
2903
2904&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2905
2906&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2907Stacking bagged and dagged models
2908&lt;/p&gt;&lt;/div&gt;
2909
2910
2911
2912&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2913
2914&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2915Kai Ming Ting, I.H. Witten
2916&lt;/p&gt;&lt;/div&gt;
2917
2918
2919
2920&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2921
2922&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2923In this paper, we investigate the method of &lt;i&gt;stacked generalization&lt;/i&gt; in combining models derived from different subsets of a training dataset by a single learning algorithm, as well as different algorithms. The simplest way to combine predictions from competing models is majority vote, and the effect of the sampling regime used to generate training subsets has already been studied in this context-when bootstrap samples are used the method is called &lt;i&gt;bagging&lt;/i&gt;, and for disjoint samples we call it &lt;i&gt;dagging&lt;/i&gt;. This paper extends these studies to stacked generalization, where a learning algorithm is employed to combine the models. This yields new methods dubbed &lt;i&gt;bag-stacking&lt;/i&gt; and &lt;i&gt;dag-stacking&lt;/i&gt;.
2924&lt;/p&gt;&lt;/div&gt;
2925
2926
2927
2928&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2929
2930&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2931
2932&lt;/p&gt;&lt;/div&gt;
2933
2934
2935
2936&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2937
2938&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2939We demonstrate that bag-stacking and dag-stacking can be effective for classification tasks even when the training samples cover just a small fraction of the full dataset. In contrast to earlier bagging results, we show that bagging and bag-stacking work for stable as well as unstable learning algorithms, as do dagging and dag-stacking. We find that bag-stacking (dag-stacking) almost always has higher predictive accuracy than bagging (dagging), and we also show that bag-stacking models derived using two different algorithms is more effective than bagging.
2940&lt;/p&gt;&lt;/div&gt;
2941
2942
2943
2944&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2945
2946&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2947
2948&lt;/p&gt;&lt;/div&gt;
2949
2950
2951
2952&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2953
2954&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2955
2956&lt;/p&gt;&lt;/div&gt;
2957
2958
2959
2960&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2961
2962&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
296397/10
2964&lt;/p&gt;&lt;/div&gt;
2965
2966
2967
2968&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2969
2970&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2971Extracting text from Postscript
2972&lt;/p&gt;&lt;/div&gt;
2973
2974
2975
2976&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2977
2978&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2979Craig Nevill-Manning, Todd Reed, Ian H. Witten
2980&lt;/p&gt;&lt;/div&gt;
2981
2982
2983
2984&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2985
2986&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2987We show how to extract plain text from PostScript files. A textual scan is inadequate because PostScript interpreters can generate characters on the page that do not appear in the source file. Furthermore, word and line breaks are implicit in the graphical rendition, and must be inferred from the positioning of word fragments. We present a robust technique for extracting text and recognizing words and paragraphs. The method uses a standard PostScript interpreter but redefines several PostScript operators, and simple heuristics are employed to locate word and line breaks. The scheme has been used to create a full-text index, and plain-text versions, of 40,000 technical reports (34 Gbyte of PostScript). Other text-extraction systems are reviewed: none offer the same combination of robustness and simplicity.
2988&lt;/p&gt;&lt;/div&gt;
2989
2990
2991
2992&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
2993
2994&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
2995
2996&lt;/p&gt;&lt;/div&gt;
2997
2998
2999
3000&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3001
3002&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3003
3004&lt;/p&gt;&lt;/div&gt;
3005
3006
3007
3008&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3009
3010&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
301197/11
3012&lt;/p&gt;&lt;/div&gt;
3013
3014
3015
3016&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3017
3018&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3019Gathering and indexing rich fragments of the World Wide Web
3020&lt;/p&gt;&lt;/div&gt;
3021
3022
3023
3024&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3025
3026&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3027Geoffrey Holmes, William J Rogers
3028&lt;/p&gt;&lt;/div&gt;
3029
3030
3031
3032&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3033
3034&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3035While the World Wide Web (WWW) is an attractive option as a resource for teaching and research it does have some undesirable features. The cost of allowing students unlimited access can be high-both in money and time; students may become addicted to 'surfing' the web-exploring purely for entertainment-and jeopardise their studies. Students are likely to discover undesirable material because large scale search engines index sites regardless of their merit. Finally, the explosive growth of WWW usage means that servers and networks are often overloaded, to the extent that a student may gain a very negative view of the technology.
3036&lt;/p&gt;&lt;/div&gt;
3037
3038
3039
3040&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3041
3042&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3043
3044&lt;/p&gt;&lt;/div&gt;
3045
3046
3047
3048&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3049
3050&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3051We have developed a piece of software which attempts to address these issues by capturing rich fragments of the WWW onto local storage media. It is possible to put a collection onto CD ROM, providing portability and inexpensive storage. This enables the presentation of the WWW to distance learning students, who do not have internet access. The software interfaces to standard, commonly available web browsers, acting as a proxy server to the files stored on the local media, and provides a search engine giving full text searching capability within the collection.
3052&lt;/p&gt;&lt;/div&gt;
3053
3054
3055
3056&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3057
3058&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3059
3060&lt;/p&gt;&lt;/div&gt;
3061
3062
3063
3064&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3065
3066&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3067
3068&lt;/p&gt;&lt;/div&gt;
3069
3070
3071
3072&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3073
3074&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3075
3076&lt;/p&gt;&lt;/div&gt;
3077
3078
3079
3080&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3081
3082&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
308397/12
3084&lt;/p&gt;&lt;/div&gt;
3085
3086
3087
3088&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3089
3090&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3091Using model trees for classification
3092&lt;/p&gt;&lt;/div&gt;
3093
3094
3095
3096&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3097
3098&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3099Eibe Frank, Yong Wang, Stuart Inglis, Geoffrey Holmes, Ian H. Witten
3100&lt;/p&gt;&lt;/div&gt;
3101
3102
3103
3104&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3105
3106&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3107Model trees, which are a type of decision tree with linear regression functions at the leaves, form the basis of a recent successful technique for predicting continuous numeric values. They can be applied to classification problems by employing a standard method of transforming a classification problem into a problem of function approximation. Surprisingly, using this simple transformation the model tree inducer M5', based on Quinlan's M5, generates more accurate classifiers than the state-of-the-art decision tree learner C5.0, particularly when most of the attributes are numeric.
3108&lt;/p&gt;&lt;/div&gt;
3109
3110
3111
3112&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3113
3114&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3115
3116&lt;/p&gt;&lt;/div&gt;
3117
3118
3119
3120&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3121
3122&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3123
3124&lt;/p&gt;&lt;/div&gt;
3125
3126
3127
3128&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3129
3130&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
313197/13
3132&lt;/p&gt;&lt;/div&gt;
3133
3134
3135
3136&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3137
3138&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3139Discovering inter-attribute relationships
3140&lt;/p&gt;&lt;/div&gt;
3141
3142
3143
3144&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3145
3146&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3147Geoffrey Holmes
3148&lt;/p&gt;&lt;/div&gt;
3149
3150
3151
3152&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3153
3154&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3155It is important to discover relationships between attributes being used to predict a class attribute in supervised learning situations for two reasons. First, any such relationship will be potentially interesting to the provider of a dataset in its own right. Second, it would simplify a learning algorithm's search space, and the related irrelevant feature and subset selection problem, if the relationships were removed from datasets ahead of learning. An algorithm to discover such relationships is presented in this paper. The algorithm is described and a surprising number of inter-attribute relationships are discovered in datasets from the University of California at Irvine (UCI) repository.
3156&lt;/p&gt;&lt;/div&gt;
3157
3158
3159
3160&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3161
3162&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3163
3164&lt;/p&gt;&lt;/div&gt;
3165
3166
3167
3168&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3169
3170&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3171
3172&lt;/p&gt;&lt;/div&gt;
3173
3174
3175
3176&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3177
3178&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
317997/14
3180&lt;/p&gt;&lt;/div&gt;
3181
3182
3183
3184&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3185
3186&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3187Learning from batched data: model combination vs data combination
3188&lt;/p&gt;&lt;/div&gt;
3189
3190
3191
3192&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3193
3194&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3195Kai Ming Ting, Boon Toh Low, Ian H. Witten
3196&lt;/p&gt;&lt;/div&gt;
3197
3198
3199
3200&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3201
3202&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3203When presented with multiple batches of data, one can either combine them into a single batch before applying a machine learning procedure or learn from each batch independently and combine the resulting models. The former procedure, data combination, is straightforward; this paper investigates the latter, model combination. Given an appropriate combination method, one might expect model combination to prove superior when the data in each batch was obtained under somewhat different conditions or when different learning algorithms were used on the batches. Empirical results show that model combination often outperforms data combination even when the batches are drawn randomly from a single source of data and the same learning method is used on each. Moreover, this is not just an artifact of one particular method of combining models: it occurs with several different combination methods.
3204&lt;/p&gt;&lt;/div&gt;
3205
3206
3207
3208&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3209
3210&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3211
3212&lt;/p&gt;&lt;/div&gt;
3213
3214
3215
3216&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3217
3218&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3219We relate this phenomenon to the learning curve of the classifiers being used. Early in the learning process when the learning curve is steep there is much to gain from data combination, but later when it becomes shallow there is less to gain and model combination achieves a greater reduction in variance and hence a lower error rate.
3220&lt;/p&gt;&lt;/div&gt;
3221
3222
3223
3224&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3225
3226&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3227
3228&lt;/p&gt;&lt;/div&gt;
3229
3230
3231
3232&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3233
3234&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3235The practical implication of these results is that one should consider using model combination rather than data combination, especially when multiple batches of data for the same task are readily available. It is often superior even when the batches are drawn randomly from a single sample, and we expect its advantage to increase if genuine statistical differences between the batches exist.
3236&lt;/p&gt;&lt;/div&gt;
3237
3238
3239
3240&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3241
3242&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3243
3244&lt;/p&gt;&lt;/div&gt;
3245
3246
3247
3248&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3249
3250&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3251
3252&lt;/p&gt;&lt;/div&gt;
3253
3254
3255
3256&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3257
3258&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
325997/15
3260&lt;/p&gt;&lt;/div&gt;
3261
3262
3263
3264&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3265
3266&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3267Information seeking retrieval, reading and storing behaviour of library users
3268&lt;/p&gt;&lt;/div&gt;
3269
3270
3271
3272&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3273
3274&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3275Turner K.
3276&lt;/p&gt;&lt;/div&gt;
3277
3278
3279
3280&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3281
3282&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3283In the interest of digital libraries, it is advisable that designers be aware of the potential behaviour of the users of such a system. There are two distinct parts under investigation, the interaction between traditional libraries involving the seeking and retrieval of relevant material, and the reading and storage behaviours ensuing. Through this analysis, the findings could be incorporated into digital library facilities. There has been copious amounts of research on information seeking leading to the development of behavioural models to describe the process. Often research on the information seeking practices of individuals is based on the task and field of study. The information seeking model, presented by Ellis et al. (1993), characterises the format of this study where it is used to compare various research on the information seeking practices of groups of people (from academics to professionals). It is found that, although researchers do make use of library facilities, they tend to rely heavily on their own collections and primarily use the library as a source for previously identified information, browsing and interloan. It was found that there are significant differences in user behaviour between the groups analysed. When looking at the reading and storage of material it was hard to draw conclusions, due to the lack of substantial research and information on the topic. However, through the use of reading strategies, a general idea on how readers behave can be developed. Designers of digital libraries can benefit from the guidelines presented here to better understand their audience.
3284&lt;/p&gt;&lt;/div&gt;
3285
3286
3287
3288&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3289
3290&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3291
3292&lt;/p&gt;&lt;/div&gt;
3293
3294
3295
3296&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3297
3298&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3299
3300&lt;/p&gt;&lt;/div&gt;
3301
3302
3303
3304&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3305
3306&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
330797/16
3308&lt;/p&gt;&lt;/div&gt;
3309
3310
3311
3312&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3313
3314&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3315Proceeding of the INTERACT97 Combined Workshop on CSCW in HCI-Worldwide
3316&lt;/p&gt;&lt;/div&gt;
3317
3318
3319
3320&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3321
3322&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3323Matthias Rauterberg, Lars Oestreicher, John Grundy
3324&lt;/p&gt;&lt;/div&gt;
3325
3326
3327
3328&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3329
3330&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3331This is the proceedings for the INTERACT97 combined workshop on &amp;ldquo;CSCW in HCI-worldwide&amp;rdquo;. The position papers in this proceedings are those selected from topics relating to HCI community development worldwide and to CSCW issues. Originally these were to be two separate INTERACT workshops, but were combined to ensure sufficient participation for a combined workshop to run.
3332&lt;/p&gt;&lt;/div&gt;
3333
3334
3335
3336&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3337
3338&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3339
3340&lt;/p&gt;&lt;/div&gt;
3341
3342
3343
3344&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3345
3346&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3347The combined workshop has been split into two separate sessions to run in the morning of July 15&lt;sup&gt;th&lt;/sup&gt;, Sydney, Australia. One to discuss the issues relating to the position papers focusing on general CSCW systems, the other to the development of HCI communities in a worldwide context. The CSCW session uses as a case study a proposed groupware tool for facilitating the development of an HCI database with a worldwide geographical distribution. The HCI community session focuses on developing the content for such a database, in order for it to foster the continued development of HCI communities. The afternoon session of the combined workshop involves a joint discussion of the case study groupware tool, in terms of its content and likely groupware facilities.
3348&lt;/p&gt;&lt;/div&gt;
3349
3350
3351
3352&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3353
3354&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3355
3356&lt;/p&gt;&lt;/div&gt;
3357
3358
3359
3360&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3361
3362&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3363The position papers have been grouped into those focusing on HCI communities and hence content issues for a groupware database, and those focusing on CSCW and groupware issues, and hence likely groupware support in the proposed HCI database/collaboration tools. We hope that you find the position papers in this proceedings offer a wide range of interesting reports of HCI community development worldwide, leading CSCW system research, and that a groupware tool supporting aspects of a worldwide HCI database can draw upon the varied work reported.
3364&lt;/p&gt;&lt;/div&gt;
3365
3366
3367
3368&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3369
3370&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3371
3372&lt;/p&gt;&lt;/div&gt;
3373
3374
3375
3376&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3377
3378&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3379
3380&lt;/p&gt;&lt;/div&gt;
3381
3382
3383
3384&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3385
3386&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3387
3388&lt;/p&gt;&lt;/div&gt;
3389
3390
3391
3392&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3393
3394&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
339597/17
3396&lt;/p&gt;&lt;/div&gt;
3397
3398
3399
3400&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3401
3402&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3403Internationalising a spreadsheet for Pacific Basin languages
3404&lt;/p&gt;&lt;/div&gt;
3405
3406
3407
3408&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3409
3410&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3411Robert Barbour, Alvin Yeo
3412&lt;/p&gt;&lt;/div&gt;
3413
3414
3415
3416&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3417
3418&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3419As people trade and engage in commerce, an economically dominant culture tends to migrate language into other recently contacted cultures. Information technology (IT) can accelerate enculturation and promote the expansion of western hegemony in IT. Equally, IT can present a culturally appropriate interface to the user that promotes the preservation of culture and language with very little additional effort. In this paper a spreadsheet is internationalised to accept languages from the Latin-1 character set such as English, Maori and Bahasa Melayu (Malaysia's national language). A technique that allows a non-programmer to add a new language to the spreadsheet is described. The technique could also be used to internationalise other software at the point of design by following the steps we outline.
3420&lt;/p&gt;&lt;/div&gt;
3421
3422
3423
3424&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3425
3426&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3427
3428&lt;/p&gt;&lt;/div&gt;
3429
3430
3431
3432&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3433
3434&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3435
3436&lt;/p&gt;&lt;/div&gt;
3437
3438
3439
3440&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3441
3442&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3443
3444&lt;/p&gt;&lt;/div&gt;
3445
3446
3447
3448&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3449
3450&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
345197/18
3452&lt;/p&gt;&lt;/div&gt;
3453
3454
3455
3456&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3457
3458&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3459Localising a spreadsheet: an Iban example
3460&lt;/p&gt;&lt;/div&gt;
3461
3462
3463
3464&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3465
3466&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3467Alvin Yeo, Robert Barbour
3468&lt;/p&gt;&lt;/div&gt;
3469
3470
3471
3472&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3473
3474&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3475Presently, there is little localisation of software to smaller cultures if it is not economically viable. We believe software should also be localised to the languages of small cultures in order to sustain and preserve these small cultures. As an example, we localised a spreadsheet from English to Iban. The process in which we carried out the localisation can be used as a framework for the localisation of software to languages of small ethnic minorities. Some problems faced during the localisation process are also discussed.
3476&lt;/p&gt;&lt;/div&gt;
3477
3478
3479
3480&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3481
3482&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3483
3484&lt;/p&gt;&lt;/div&gt;
3485
3486
3487
3488&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3489
3490&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3491
3492&lt;/p&gt;&lt;/div&gt;
3493
3494
3495
3496&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3497
3498&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3499
3500&lt;/p&gt;&lt;/div&gt;
3501
3502
3503
3504&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3505
3506&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
350797/19
3508&lt;/p&gt;&lt;/div&gt;
3509
3510
3511
3512&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3513
3514&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3515Strategies of internationalisation and localisation: a postmodernist/s perspective
3516&lt;/p&gt;&lt;/div&gt;
3517
3518
3519
3520&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3521
3522&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3523Alvin Yeo, Robert Barbour
3524&lt;/p&gt;&lt;/div&gt;
3525
3526
3527
3528&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3529
3530&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3531Many software companies today are developing software not only for local consumption but for the rest of the world. We introduce the concepts of internationalisation and localisation and discuss some techniques using these processes. An examination of postmodern critique with respect to the software industry is also reported. In addition, we also feature our proposed internationalisation technique that was inspired by taking into account the researches of postmodern philosophers and mathematicians. As illustrated in our prototype, the technique empowers non-programmers to localise their own software. Further development of the technique and its implications on user interfaces and the future of software internationalisation and localisation are discussed.
3532&lt;/p&gt;&lt;/div&gt;
3533
3534
3535
3536&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3537
3538&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3539
3540&lt;/p&gt;&lt;/div&gt;
3541
3542
3543
3544&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3545
3546&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3547
3548&lt;/p&gt;&lt;/div&gt;
3549
3550
3551
3552&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3553
3554&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
355597/20
3556&lt;/p&gt;&lt;/div&gt;
3557
3558
3559
3560&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3561
3562&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3563Language use in software
3564&lt;/p&gt;&lt;/div&gt;
3565
3566
3567
3568&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3569
3570&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3571Alvin Yeo, Robert Barbour
3572&lt;/p&gt;&lt;/div&gt;
3573
3574
3575
3576&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3577
3578&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3579Many of the popular software we use today are in English. Very few software applications are available in minority languages. Besides economic goals, we justify why software should be made available to smaller cultures. Furthermore, there is evidence that people learn and progress faster in software in their mother tongue (Griffiths et at, 1994) (Krock, 1996). We hypothesise that experienced users of English spreadsheet can easily migrate to a spreadsheet in their native tongue i.e. Bahasa Melayu (Malaysia's national language). Observations made in the study suggest that the native speakers of Bahasa Melayu had difficulties with the Bahasa Melayu interface. The subjects' main difficulty was their unfamiliarity with computing terminology in Bahasa Melayu. We present possible strategies to increase the use of Bahasa Melayu in IT. These strategies may also be used to promote the use of other minority languages in IT.
3580&lt;/p&gt;&lt;/div&gt;
3581
3582
3583
3584&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3585
3586&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3587
3588&lt;/p&gt;&lt;/div&gt;
3589
3590
3591
3592&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3593
3594&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3595
3596&lt;/p&gt;&lt;/div&gt;
3597
3598
3599
3600&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3601
3602&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3603
3604&lt;/p&gt;&lt;/div&gt;
3605
3606
3607
3608&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3609
3610&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
361197/21
3612&lt;/p&gt;&lt;/div&gt;
3613
3614
3615
3616&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3617
3618&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3619Usability testing: a Malaysian study
3620&lt;/p&gt;&lt;/div&gt;
3621
3622
3623
3624&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3625
3626&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3627Alvin Yeo, Robert Barbour, Mark Apperley
3628&lt;/p&gt;&lt;/div&gt;
3629
3630
3631
3632&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3633
3634&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3635An exploratory study of software assessment techniques is conducted in Malaysia. Subjects in the study comprised staff members of a Malaysian university with a high Information Technology (IT) presence. The subjects assessed a spreadsheet tool with a Bahasa Melayu (Malaysia's national language) interface. Software evaluation techniques used include the think aloud method, interviews and the System Usability Scale. The responses in the various techniques used are reported and initial results indicate idiosyncratic behaviour of Malaysian subjects. The implications of the findings are also discussed.
3636&lt;/p&gt;&lt;/div&gt;
3637
3638
3639
3640&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3641
3642&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3643
3644&lt;/p&gt;&lt;/div&gt;
3645
3646
3647
3648&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3649
3650&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3651
3652&lt;/p&gt;&lt;/div&gt;
3653
3654
3655
3656&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3657
3658&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3659
3660&lt;/p&gt;&lt;/div&gt;
3661
3662
3663
3664&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3665
3666&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
366797/22
3668&lt;/p&gt;&lt;/div&gt;
3669
3670
3671
3672&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3673
3674&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3675Inducing cost-sensitive trees via instance-weighting
3676&lt;/p&gt;&lt;/div&gt;
3677
3678
3679
3680&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3681
3682&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3683Kai Ming Ting
3684&lt;/p&gt;&lt;/div&gt;
3685
3686
3687
3688&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3689
3690&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3691We introduce an instance-weighting method to induce cost-sensitive trees in this paper. It is a generalization of the standard tree induction process where only the initial instance weights determine the type of tree (i.e., minimum error trees or minimum cost trees) to be induced. We demonstrate that it can be easily adopted to an existing tree learning algorithm.
3692&lt;/p&gt;&lt;/div&gt;
3693
3694
3695
3696&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3697
3698&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3699
3700&lt;/p&gt;&lt;/div&gt;
3701
3702
3703
3704&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3705
3706&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3707Previous research gave insufficient evidence to support the fact that the greedy divide-and-conquer algorithm can effectively induce a truly cost-sensitive tree directly from the training data. We provide this empirical evidence in this paper. The algorithm employing the instance-weighting method is found to be comparable to or better than both C4.5 and C5 in terms of total misclassification costs, tree size and the number of high cost errors. The instance-weighting method is also simpler and more effective in implementation than a method based on altered priors.
3708&lt;/p&gt;&lt;/div&gt;
3709
3710
3711
3712&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3713
3714&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3715
3716&lt;/p&gt;&lt;/div&gt;
3717
3718
3719
3720&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3721
3722&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3723
3724&lt;/p&gt;&lt;/div&gt;
3725
3726
3727
3728&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3729
3730&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
373197/23
3732&lt;/p&gt;&lt;/div&gt;
3733
3734
3735
3736&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3737
3738&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3739Fast convergence with a greedy tag-phrase dictionary
3740&lt;/p&gt;&lt;/div&gt;
3741
3742
3743
3744&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3745
3746&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3747Ross Peeters, Tony C. Smith
3748&lt;/p&gt;&lt;/div&gt;
3749
3750
3751
3752&lt;p&gt;&lt;div name=&quot;Body Text&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3753
3754&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3755The best general-purpose compression schemes make their gains by estimating a probability distribution over all possible next symbols given the context established by some number of previous symbols. Such context models typically obtain good compression results for plain text by taking advantage of regularities in character sequences. Frequent words and syllables can be incorporated into the model quickly and thereafter used for reasonably accurate prediction. However, the precise context in which frequent patterns emerge is often extremely varied, and each new word or phrase immediately introduces new contexts which can adversely affect the compression rate
3756&lt;/p&gt;&lt;/div&gt;
3757
3758
3759
3760&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3761
3762&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3763
3764&lt;/p&gt;&lt;/div&gt;
3765
3766
3767
3768&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3769
3770&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3771A great deal of the structural regularity in a natural language is given rather more by properties of its grammar than by the orthographic transcription of its phonology. This implies that access to a grammatical abstraction might lead to good compression. While grammatical models have been used successfully for compressing computer programs [4], grammar-based compression of plain text has received little attention, primarily because of the difficulties associated with constructing a suitable natural language grammar. But even without a precise formulation of the syntax of a language, there is a linguistic abstraction which is easily accessed and which demonstrates a high degree of regularity which can be exploited for compression purposes-namely, lexical categories.
3772&lt;/p&gt;&lt;/div&gt;
3773
3774
3775
3776&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3777
3778&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3779
3780&lt;/p&gt;&lt;/div&gt;
3781
3782
3783
3784&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3785
3786&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3787
3788&lt;/p&gt;&lt;/div&gt;
3789
3790
3791
3792&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3793
3794&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
379597/24
3796&lt;/p&gt;&lt;/div&gt;
3797
3798
3799
3800&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3801
3802&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3803Tag based models of English text
3804&lt;/p&gt;&lt;/div&gt;
3805
3806
3807
3808&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3809
3810&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3811W. J. Teahan, John G. Cleary
3812&lt;/p&gt;&lt;/div&gt;
3813
3814
3815
3816&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3817
3818&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3819The problem of compressing English text is important both because of the ubiquity of English as a target for compression and because of the light that compression can shed on the structure of English. English text is examined in conjunction with additional information about the parts of speech of each word in the text (these are referred to as &amp;ldquo;tags&amp;rdquo;). It is shown that the tags plus the text can be compressed more than the text alone. Essentially the tags can be compressed for nothing or even a small net saving in size. A comparison is made of a number of different ways of integrating compression of tags and text using an escape mechanism similar to PPM. These are also compared with standard word based and character based compression programs. The result is that the tag character and word based schemes always outperform the character based schemes. Overall, the tag based schemes outperform the word based schemes. We conclude by conjecturing that tags chosen for compression rather than linguistic purposes would perform even better.
3820&lt;/p&gt;&lt;/div&gt;
3821
3822
3823
3824&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3825
3826&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3827
3828&lt;/p&gt;&lt;/div&gt;
3829
3830
3831
3832&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3833
3834&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3835
3836&lt;/p&gt;&lt;/div&gt;
3837
3838
3839
3840&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3841
3842&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
384397/25
3844&lt;/p&gt;&lt;/div&gt;
3845
3846
3847
3848&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3849
3850&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3851Musical image compression
3852&lt;/p&gt;&lt;/div&gt;
3853
3854
3855
3856&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3857
3858&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3859David Bainbridge, Stuart Inglis
3860&lt;/p&gt;&lt;/div&gt;
3861
3862
3863
3864&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3865
3866&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3867Optical music recognition aims to convert the vast repositories of sheet music in the world into an on-line digital format [Bai97]. In the near future it will be possible to assimilate music into digital libraries and users will be able to perform searches based on a sung melody in addition to typical text-based searching [MSW+96]. An important requirement for such a system is the ability to reproduce the original score as accurately as possible. Due to the huge amount of sheet music available, the efficient storage of musical images is an important topic of study.
3868&lt;/p&gt;&lt;/div&gt;
3869
3870
3871
3872&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3873
3874&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3875
3876&lt;/p&gt;&lt;/div&gt;
3877
3878
3879
3880&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3881
3882&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3883This paper investigates whether the &amp;ldquo;knowledge&amp;rdquo; extracted from the optical music recognition (OMR) process can be exploited to gain higher compression than the JBIG international standard for bi-level image compression. We present a hybrid approach where the primitive shapes of music extracted by the optical music recognition process-note heads, note stems, staff lines and so forth-are fed into a graphical symbol based compression scheme originally designed for images containing mainly printed text. Using this hybrid approach the average compression rate for a single page is improved by 3.5% over JBIG. When multiple pages with similar typography are processed in sequence, the file size is decreased by 4-8%.
3884&lt;/p&gt;&lt;/div&gt;
3885
3886
3887
3888&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3889
3890&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3891
3892&lt;/p&gt;&lt;/div&gt;
3893
3894
3895
3896&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3897
3898&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3899Section 2 presents the relevant background to both optical music recognition and textual image compression. Section 3 describes the experiments performed on 66 test images, outlining the combinations of parameters that were examined to give the best results. The initial results and refinements are presented in Section 4, and we conclude in the last section by summarizing the findings of this work.
3900&lt;/p&gt;&lt;/div&gt;
3901
3902
3903
3904&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3905
3906&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3907
3908&lt;/p&gt;&lt;/div&gt;
3909
3910
3911
3912&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3913
3914&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3915
3916&lt;/p&gt;&lt;/div&gt;
3917
3918
3919
3920&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3921
3922&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3923
3924&lt;/p&gt;&lt;/div&gt;
3925
3926
3927
3928&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3929
3930&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
393197/26
3932&lt;/p&gt;&lt;/div&gt;
3933
3934
3935
3936&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3937
3938&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3939Correcting English text using PPM models
3940&lt;/p&gt;&lt;/div&gt;
3941
3942
3943
3944&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3945
3946&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3947W. J. Teahan, S. Inglis, J. G. Cleary, G. Holmes
3948&lt;/p&gt;&lt;/div&gt;
3949
3950
3951
3952&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3953
3954&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3955An essential component of many applications in natural language processing is a language modeler able to correct errors in the text being processed. For optical character recognition (OCR), poor scanning quality or extraneous pixels in the image may cause one or more characters to be mis-recognized; while for spelling correction, two characters may be transposed, or a character may be inadvertently inserted or missed out.
3956&lt;/p&gt;&lt;/div&gt;
3957
3958
3959
3960&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3961
3962&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3963
3964&lt;/p&gt;&lt;/div&gt;
3965
3966
3967
3968&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3969
3970&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3971This paper describes a method for correcting English text using a PPM model. A method that segments words in English text is introduced and is shown to be a significant improvement over previously used methods. A similar technique is also applied as a post-processing stage after pages have been recognized by a state-of-the-art commercial OCR system. We show that the accuracy of the OCR system can be increased from 95.9% to 96.6%, a decrease of about 10 errors per page.
3972&lt;/p&gt;&lt;/div&gt;
3973
3974
3975
3976&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3977
3978&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3979
3980&lt;/p&gt;&lt;/div&gt;
3981
3982
3983
3984&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3985
3986&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3987
3988&lt;/p&gt;&lt;/div&gt;
3989
3990
3991
3992&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
3993
3994&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
3995
3996&lt;/p&gt;&lt;/div&gt;
3997
3998
3999
4000&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4001
4002&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4003
4004&lt;/p&gt;&lt;/div&gt;
4005
4006
4007
4008&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4009
4010&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
401197/27
4012&lt;/p&gt;&lt;/div&gt;
4013
4014
4015
4016&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4017
4018&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4019Constraints on parallelism beyond 10 instructions per cycle
4020&lt;/p&gt;&lt;/div&gt;
4021
4022
4023
4024&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4025
4026&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4027John G. Cleary, Richard H. Littin, J. A. David McWha, Murray W. Pearson
4028&lt;/p&gt;&lt;/div&gt;
4029
4030
4031
4032&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4033
4034&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4035The problem of extracting Instruction Level Parallelism at levels of 10 instructions per clock and higher is considered. Two different architectures which use speculation on memory accesses to achieve this level of performance are reviewed. It is pointed out that while this form of speculation gives high potential parallelism it is necessary to retain execution state so that incorrect speculation can be detected and subsequently squashed. Simulation results show that the space to store such state is a critical resource in obtaining good speedup. To make good use of the space it is essential that state be stored efficiently and that it be retired as soon as possible. A number of techniques for extracting the best usage from the available state storage are introduced.
4036&lt;/p&gt;&lt;/div&gt;
4037
4038
4039
4040&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4041
4042&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4043
4044&lt;/p&gt;&lt;/div&gt;
4045
4046
4047
4048&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4049
4050&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4051
4052&lt;/p&gt;&lt;/div&gt;
4053
4054
4055
4056&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4057
4058&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4059
4060&lt;/p&gt;&lt;/div&gt;
4061
4062
4063
4064&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4065
4066&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
406797/28
4068&lt;/p&gt;&lt;/div&gt;
4069
4070
4071
4072&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4073
4074&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4075Effects of re-ordered memory operations on parallelism
4076&lt;/p&gt;&lt;/div&gt;
4077
4078
4079
4080&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4081
4082&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4083Richard H. Littin, John G. Cleary
4084&lt;/p&gt;&lt;/div&gt;
4085
4086
4087
4088&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4089
4090&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4091The performance effect of permitting different memory operations to be re-ordered is examined. The available parallelism is computed using a machine code simulator. A range of possible restrictions on the re-ordering of memory operations is considered: from the purely sequential case where no re-ordering is permitted; to the completely permissive one where memory operations may occur in any order so that the parallelism is restricted only by data dependencies. A general conclusion is drawn that to reliably obtain parallelism beyond 10 instructions per clock will require an ability to re-order all memory instructions. A brief description of a feasible architecture capable of this is given.
4092&lt;/p&gt;&lt;/div&gt;
4093
4094
4095
4096&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4097
4098&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4099
4100&lt;/p&gt;&lt;/div&gt;
4101
4102
4103
4104&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4105
4106&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4107
4108&lt;/p&gt;&lt;/div&gt;
4109
4110
4111
4112&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4113
4114&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4115
4116&lt;/p&gt;&lt;/div&gt;
4117
4118
4119
4120&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot;margin: 0.00mm -0.14mm 0.00mm 0.00mm; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4121
4122&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
412397/29
4124&lt;/p&gt;&lt;/div&gt;
4125
4126
4127
4128&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4129
4130&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4131OZCHI'96 Industry Session: Sixth Australian Conference on Human-Computer Interaction
4132&lt;/p&gt;&lt;/div&gt;
4133
4134
4135
4136&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4137
4138&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4139Edited by Chris Phillips, Janis McKauge
4140&lt;/p&gt;&lt;/div&gt;
4141
4142
4143
4144&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4145
4146&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4147The idea for a specific industry session at OZCHI was first mooted at the 1995 conference in Wollongong, during questions following a session of short papers which happened (serendipitously) to be presented by people from industry. An animated discussion took place, most of which was about how OZCHI could be made more relevant to people in industry, be it working as usability consultants, or working within organisations either as usability professionals or as `champions of the cause'. The discussion raised more questions than answers, about the format of such as session, about the challenges of attracting industry participation, and about the best way of publishing the results. Although no real solutions were arrived at, it was enough to place an industry session on the agenda for OZCHI'96.
4148&lt;/p&gt;&lt;/div&gt;
4149
4150
4151
4152&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4153
4154&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4155
4156&lt;/p&gt;&lt;/div&gt;
4157
4158
4159
4160&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4161
4162&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4163
4164&lt;/p&gt;&lt;/div&gt;
4165
4166
4167
4168&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4169
4170&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
417197/30
4172&lt;/p&gt;&lt;/div&gt;
4173
4174
4175
4176&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4177
4178&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4179Adaptive models of English text
4180&lt;/p&gt;&lt;/div&gt;
4181
4182
4183
4184&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4185
4186&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4187W. J. Teahan, John G. Cleary
4188&lt;/p&gt;&lt;/div&gt;
4189
4190
4191
4192&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4193
4194&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4195High quality models of English text with performance approaching that of humans is important for many applications including spelling correction, speech recognition, OCR, and encryption. A number of different statistical models of English are compared with each other and with previous estimates from human subjects. It is concluded that the best current models are word based with part of speech tags. Given sufficient training text, they are able to attain performance comparable to humans.
4196&lt;/p&gt;&lt;/div&gt;
4197
4198
4199
4200&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4201
4202&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4203
4204&lt;/p&gt;&lt;/div&gt;
4205
4206
4207
4208&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4209
4210&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4211
4212&lt;/p&gt;&lt;/div&gt;
4213
4214
4215
4216&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4217
4218&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4219
4220&lt;/p&gt;&lt;/div&gt;
4221
4222
4223
4224&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4225
4226&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
422797/31
4228&lt;/p&gt;&lt;/div&gt;
4229
4230
4231
4232&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4233
4234&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4235A graphical user interface for Boolean query specification
4236&lt;/p&gt;&lt;/div&gt;
4237
4238
4239
4240&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4241
4242&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4243Steve Jones, Shona McInnes
4244&lt;/p&gt;&lt;/div&gt;
4245
4246
4247
4248&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4249
4250&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4251On-line information repositories commonly provide keyword search facilities via textual query languages based on Boolean logic. However, there is evidence to suggest that the syntactical demands of such languages can lead to user errors and adversely affect the time that it takes users to form queries. Users also face difficulties because of the conflict in semantics between AND and OR when used in Boolean logic and English language. We suggest that graphical query languages, in particular Venn-like diagrams, can alleviate the problems that users experience when forming Boolean expressions with textual languages. We describe Vquery, a Venn-diagram based user interface to the New Zealand Digital Library (NZDL). The design of Vquery has been partly motivated by analysis of NZDL usage. We found that few queries contain more than three terms, use of the intersection operator dominates and that query refinement is common. A study of the utility of Venn diagrams for query specification indicates that with little or no training users can interpret and form Venn-like diagrams which accurately correspond to Boolean expressions. The utility of Vquery is considered and directions for future work are proposed.
4252&lt;/p&gt;&lt;/div&gt;
4253
4254
4255
4256&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4257
4258&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4259
4260&lt;/p&gt;&lt;/div&gt;
4261
4262
4263
4264&lt;p&gt;&lt;div name=&quot;Normal&quot; align=&quot;left&quot; style=&quot; padding: 0.00mm 0.00mm 0.00mm 0.00mm; &quot;&gt;
4265
4266&lt;p style=&quot;text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; &quot;&gt;
4267
4268&lt;/p&gt;&lt;/div&gt;
4269
4270&lt;!--Section Ends--&gt;
4271
4272
4273
4274&lt;!--
4275&lt;hr&gt;
4276&lt;address&gt;
4277&lt;a href=&quot;_httpextlink_&amp;amp;rl=0&amp;amp;href=http:%2f%2fwvware.sourceforge.net&quot;&gt;&lt;img
4278src=&quot;_httpdocimg_/wvSmall.gif&quot; height=31 width=47
4279align=left border=0 alt=&quot;wvWare&quot;&gt;&lt;/a&gt;
4280&lt;a href=&quot;_httpextlink_&amp;amp;rl=0&amp;amp;href=http:%2f%2fvalidator.w3.org%2fcheck%2freferer&quot;&gt;&lt;img
4281src=&quot;_httpdocimg_/vh40.gif&quot; height=31 width=88
4282align=right border=0 alt=&quot;Valid HTML 4.0!&quot;&gt;&lt;/a&gt;
4283Document created with &lt;a href=&quot;_httpextlink_&amp;amp;rl=0&amp;amp;href=http:%2f%2fwvware.sourceforge.net&quot;&gt;wvWare/wvWare version 1.2.4&lt;/a&gt;&lt;br&gt;
4284&lt;/address&gt;
4285--&gt;
4286
4287
4288</Content>
4289</Section>
4290</Archive>
Note: See TracBrowser for help on using the repository browser.