| 1 | /*
|
|---|
| 2 | * This program is free software; you can redistribute it and/or modify
|
|---|
| 3 | * it under the terms of the GNU General Public License as published by
|
|---|
| 4 | * the Free Software Foundation; either version 2 of the License, or
|
|---|
| 5 | * (at your option) any later version.
|
|---|
| 6 | *
|
|---|
| 7 | * This program is distributed in the hope that it will be useful,
|
|---|
| 8 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|---|
| 9 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|---|
| 10 | * GNU General Public License for more details.
|
|---|
| 11 | *
|
|---|
| 12 | * You should have received a copy of the GNU General Public License
|
|---|
| 13 | * along with this program; if not, write to the Free Software
|
|---|
| 14 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
|---|
| 15 | */
|
|---|
| 16 |
|
|---|
| 17 | /*
|
|---|
| 18 | * DistributionClassifier.java
|
|---|
| 19 | * Copyright (C) 1999 Eibe Frank, Len Trigg
|
|---|
| 20 | *
|
|---|
| 21 | */
|
|---|
| 22 |
|
|---|
| 23 | package weka.classifiers;
|
|---|
| 24 |
|
|---|
| 25 | import weka.core.*;
|
|---|
| 26 |
|
|---|
| 27 | /**
|
|---|
| 28 | * Abstract classification model that produces (for each test instance)
|
|---|
| 29 | * an estimate of the membership in each class
|
|---|
| 30 | * (ie. a probability distribution).
|
|---|
| 31 | *
|
|---|
| 32 | * @author Eibe Frank ([email protected])
|
|---|
| 33 | * @author Len Trigg ([email protected])
|
|---|
| 34 | * @version $Revision: 8815 $
|
|---|
| 35 | */
|
|---|
| 36 | public abstract class DistributionClassifier extends Classifier {
|
|---|
| 37 |
|
|---|
| 38 | /**
|
|---|
| 39 | * Predicts the class memberships for a given instance. If
|
|---|
| 40 | * an instance is unclassified, the returned array elements
|
|---|
| 41 | * must be all zero. If the class is numeric, the array
|
|---|
| 42 | * must consist of only one element, which contains the
|
|---|
| 43 | * predicted value.
|
|---|
| 44 | *
|
|---|
| 45 | * @param instance the instance to be classified
|
|---|
| 46 | * @return an array containing the estimated membership
|
|---|
| 47 | * probabilities of the test instance in each class (this
|
|---|
| 48 | * should sum to at most 1)
|
|---|
| 49 | * @exception Exception if distribution could not be
|
|---|
| 50 | * computed successfully
|
|---|
| 51 | */
|
|---|
| 52 | public abstract double[] distributionForInstance(Instance instance)
|
|---|
| 53 | throws Exception;
|
|---|
| 54 |
|
|---|
| 55 | /**
|
|---|
| 56 | * Classifies the given test instance. The instance has to belong to a
|
|---|
| 57 | * dataset when it's being classified.
|
|---|
| 58 | *
|
|---|
| 59 | * @param instance the instance to be classified
|
|---|
| 60 | * @return the predicted most likely class for the instance or
|
|---|
| 61 | * Instance.missingValue() if no prediction is made
|
|---|
| 62 | * @exception Exception if an error occurred during the prediction
|
|---|
| 63 | */
|
|---|
| 64 | public double classifyInstance(Instance instance) throws Exception {
|
|---|
| 65 |
|
|---|
| 66 | double [] dist = distributionForInstance(instance);
|
|---|
| 67 | if (dist == null) {
|
|---|
| 68 | throw new Exception("Null distribution predicted");
|
|---|
| 69 | }
|
|---|
| 70 | switch (instance.classAttribute().type()) {
|
|---|
| 71 | case Attribute.NOMINAL:
|
|---|
| 72 | double max = 0;
|
|---|
| 73 | int maxIndex = 0;
|
|---|
| 74 |
|
|---|
| 75 | for (int i = 0; i < dist.length; i++) {
|
|---|
| 76 | if (dist[i] > max) {
|
|---|
| 77 | maxIndex = i;
|
|---|
| 78 | max = dist[i];
|
|---|
| 79 | }
|
|---|
| 80 | }
|
|---|
| 81 | if (max > 0) {
|
|---|
| 82 | return maxIndex;
|
|---|
| 83 | } else {
|
|---|
| 84 | return Instance.missingValue();
|
|---|
| 85 | }
|
|---|
| 86 | case Attribute.NUMERIC:
|
|---|
| 87 | return dist[0];
|
|---|
| 88 | default:
|
|---|
| 89 | return Instance.missingValue();
|
|---|
| 90 | }
|
|---|
| 91 | }
|
|---|
| 92 | }
|
|---|
