source: other-projects/maori-lang-detection/mongodb-data/counts_sitesWithPagesContainingMRI.json@ 33813

/*
Number of sites containing at least one sentence for which OpenNLP detected the best language = MRI

db.getCollection('Websites').find({numPagesContainingMRI: {$gt: 0}}).count()
868


Obviously, the following should be equal to that:

db.getCollection('Websites').find({ $or: [ { numPagesInMRI: { $gt: 0 } }, { numPagesContainingMRI: {$gt: 0} } ] } ).count()
868


Count of country codes for sites that have at least one page containing at least one sentence detected as MRI by OpenNLP:

db.Websites.aggregate([
    {
        $match: {
            numPagesContainingMRI: {$gt: 0}
        }
    },
    { $unwind: "$geoLocationCountryCode" },
    {
        $group: {
            _id: {$toLower: '$geoLocationCountryCode'},
            count: { $sum: 1 }
        }
    },
    { $sort : { count : -1} }
]);

*/

/* 1 */
{
    "_id" : "us",
    "count" : 486.0
}

/* 2 */
{
    "_id" : "cn",
    "count" : 114.0
}

/* 3 */
{
    "_id" : "nz",
    "count" : 89.0
}

/* 4 */
{
    "_id" : "fr",
    "count" : 36.0
}

/* 5 */
{
    "_id" : "de",
    "count" : 27.0
}

/* 6 */
{
    "_id" : "nl",
    "count" : 22.0
}

/* 7 */
{
    "_id" : "au",
    "count" : 21.0
}

/* 8 */
{
    "_id" : "ca",
    "count" : 12.0
}

/* 9 */
{
    "_id" : "dk",
    "count" : 8.0
}

/* 10 */
{
    "_id" : "es",
    "count" : 7.0
}

/* 11 */
{
    "_id" : "gb",
    "count" : 7.0
}

/* 12 */
{
    "_id" : "cz",
    "count" : 4.0
}

/* 13 */
{
    "_id" : "unknown",
    "count" : 3.0
}

/* 14 */
{
    "_id" : "at",
    "count" : 3.0
}

/* 15 */
{
    "_id" : "ro",
    "count" : 3.0
}

/* 16 */
{
    "_id" : "it",
    "count" : 3.0
}

/* 17 */
{
    "_id" : "sg",
    "count" : 2.0
}

/* 18 */
{
    "_id" : "jp",
    "count" : 2.0
}

/* 19 */
{
    "_id" : "ie",
    "count" : 2.0
}

/* 20 */
{
    "_id" : "hk",
    "count" : 2.0
}

/* 21 */
{
    "_id" : "ua",
    "count" : 2.0
}

/* 22 */
{
    "_id" : "ru",
    "count" : 2.0
}

/* 23 */
{
    "_id" : "ch",
    "count" : 2.0
}

/* 24 */
{
    "_id" : "il",
    "count" : 2.0
}

/* 25 */
{
    "_id" : "tr",
    "count" : 1.0
}

/* 26 */
{
    "_id" : "mx",
    "count" : 1.0
}

/* 27 */
{
    "_id" : "ir",
    "count" : 1.0
}

/* 28 */
{
    "_id" : "gr",
    "count" : 1.0
}

/* 29 */
{
    "_id" : "bg",
    "count" : 1.0
}

/* 30 */
{
    "_id" : "eu",
    "count" : 1.0
}

/* 31 */
{
    "_id" : "fi",
    "count" : 1.0
}