A class to detect the language of a text (supports German/English but can be extended)

langdetect.class.php

class LangDetect {
    protected $dict = array(
            'German' => array('ich','du','er','sie','es','wir','ihr','sie','der','die','das'),
            'English' => array('i','you','he','she','it','we','us','they','this', 'the', 'a')
        );
    protected $sourceText;    
    
    public function __construct($text) {
        $this->sourceText = $text;
    }
    
    public function getLanguage() {
        $maxScore = 0;
        $resultLang = 'Unknown'; //Default language as a fallback
        $scores = array();
        foreach($this->dict as $lang => $words) {
            $score = 0;
            foreach($words as $word) {
                if(($matches = preg_match_all('#(?<!\w)'.$word.'(?!\w)#is', $this->sourceText, $sub)) !== false )
                    $score += $matches;
            }
            if($score > $maxScore) {
                $maxScore = $score;
                $resultLang = $lang;
            }
        }
        return $resultLang;
    }
}