31 lines
897 B
Text
31 lines
897 B
Text
<?php
|
|
|
|
require __DIR__ . '/vendor/autoload.php';
|
|
|
|
$text = require __DIR__ . '/text.php';
|
|
$tokens = tokenize($text, \TextAnalysis\Tokenizers\PennTreeBankTokenizer::class);
|
|
$normalizedTokens = normalize_tokens($tokens, 'mb_strtolower');
|
|
$freqDist = freq_dist($normalizedTokens);
|
|
|
|
$rake = rake($normalizedTokens, 3);
|
|
$results = $rake->getKeywordScores();
|
|
|
|
|
|
$nb = naive_bayes();
|
|
$nb->train('mexican', tokenize('taco nacho enchilada burrito'));
|
|
$nb->train('american', tokenize('hamburger burger fries pop'));
|
|
$nb->train('alerte', normalize_tokens(tokenize('alerte message population exercice mairie cevezo victimes zone téléphone POI PPI'), 'mb_strtolower'));
|
|
$nb->train('publicite', tokenize('produit promotion euro'));
|
|
|
|
|
|
|
|
$prediction = $nb->predict($normalizedTokens);
|
|
|
|
|
|
var_dump(
|
|
$normalizedTokens,
|
|
ngrams($normalizedTokens),
|
|
stem($normalizedTokens),
|
|
$results,
|
|
$prediction,
|
|
);
|