function kMeans($data, $k, $normalize = false)
{
$centroids = initialiseCentroids($data, $k, $normalize = false);
$mapping = array();
while (true) {
$new_mapping = assignCentroids($data, $centroids);
foreach ($new_mapping as $documentID => $centroidID) {
if (!isset($mapping[$documentID]) || $centroidID != $mapping[$documentID]) {
$mapping = $new_mapping;
break;
} else {
return formatResults($mapping, $data, $centroids);
}
}
$centroids = updateCentroids($mapping, $data, $k);
}
}
function kMeans($normalisedTfidf, $numClusters)
{
// Initialise centroids ( with key = term and value = tf-idf )
$centroids = initialiseCentroids($normalisedTfidf, $numClusters);
$mapDocToCentroid = array();
$maxIterations = 10;
while (true) {
// Assign documents to centroids
$newMapDocToCentroid = assignCentroids($normalisedTfidf, $centroids);
// Stopping condition: when the centroid stops changing
$changed = false;
// Resulting array has docID for key and centroidID for values
foreach ($newMapDocToCentroid as $docID => $centroidID) {
// If centroid for a doc in new mapping is not same as that of old mapping, use the new mapping
if (!isset($mapDocToCentroid[$docID]) || $centroidID != $mapDocToCentroid[$docID]) {
$mapDocToCentroid = $newMapDocToCentroid;
$changed = true;
break;
}
}
// Decrease iterations (which were set as alternative stopping condition)
$maxIterations--;
// Check status of assignment and return the array with docs mapped to centroids
if (!$changed || $maxIterations == 0) {
return $mapDocToCentroid;
}
// Update the centroids
$centroids = updateCentroids($mapDocToCentroid, $normalisedTfidf, $numClusters);
}
}