public function getFeatureArray($class, DocumentInterface $doc)
{
$tokens = $doc->getDocumentData();
$tokens = array_count_values($tokens);
foreach ($tokens as $tok => &$v) {
$v = min($v, 4);
}
return $tokens;
}
public function classify(array $classes, DocumentInterface $d)
{
list($token, $before, $after) = $d->getDocumentData();
$dotcnt = count(explode('.', $token)) - 1;
$lastdot = substr($token, -1) == '.';
if (!$lastdot) {
// assume that all sentences end in full stops
return 'O';
}
if ($dotcnt > 1) {
// to catch some naive abbreviations (e.g.: U.S.A.)
return 'O';
}
return 'EOW';
}
/**
* For use with TokensDocument mostly.
* Simply return the data as
* features. Could contain duplicates (a feature firing twice in
* for a signle document).
*
* @param string $class
* The class for which we are calculating features
* @param DocumentInterface $d
* The document to calculate features for.
* @return array
*/
public function getFeatureArray($class, DocumentInterface $d)
{
return $d->getDocumentData();
}
public function classify(array $classes, DocumentInterface $d)
{
return $classes[$d->getDocumentData() % count($classes)];
}
