PHP改进计算字符串相似度的函数similar_text()、levenshtein()

similar_text()中文汉字版

代码如下:
     <?php 
     //拆分字符串 
     function split_str($str) { 
       preg_match_all("/./u", $str, $arr); 
       return $arr[0]; 
     } 
      
     //相似度检测 
     function similar_text_cn($str1, $str2) { 
       $arr_1 = array_unique(split_str($str1)); 
       $arr_2 = array_unique(split_str($str2)); 
       $similarity = count($arr_2) - count(array_diff($arr_2, $arr_1)); 
        
       return $similarity; 
     }  

levenshtein()中文汉字版
 

代码如下:
     <?php 
     //拆分字符串 
     function mbStringToArray($string, $encoding = 'UTF-8') { 
         $arrayResult = array(); 
         while ($iLen = mb_strlen($string, $encoding)) { 
             array_push($arrayResult, mb_substr($string, 0, 1, $encoding)); 
             $string = mb_substr($string, 1, $iLen, $encoding); 
         } 
         return $arrayResult; 
     } 
     //编辑距离 
     function levenshtein_cn($str1, $str2, $costReplace = 1, $encoding = 'UTF-8') { 
         $count_same_letter = 0; 
         $d = array(); 
         $mb_len1 = mb_strlen($str1, $encoding); 
         $mb_len2 = mb_strlen($str2, $encoding); 
         $mb_str1 = mbStringToArray($str1, $encoding); 
         $mb_str2 = mbStringToArray($str2, $encoding); 
         for ($i1 = 0; $i1 <= $mb_len1; $i1++) { 
             $d[$i1] = array(); 
             $d[$i1][0] = $i1;