PHP는 문자열 유사성 계산을 위한 유사_text() 및 levenshtein()_php 함수를 개선합니다.

similar_text() 한자버전

코드 복사 코드는 다음과 같습니다.

//문자열 분할
함수 Split_str($str) {
Preg_match_all("/./u", $str, $arr)
$arr[0] 반환
}  
                         
//유사성 감지
함수 유사한_텍스트_cn($str1, $str2) {
        $arr_1 = array_unique(split_str($str1))
        $arr_2 = array_unique(split_str($str2))
$similarity = 개수($arr_2) - 개수(array_diff($arr_2, $arr_1))
                                    
$유사성 반환
}  

levenshtein() 한자버전

코드 복사 코드는 다음과 같습니다.

          //拆分字符串 
     function mbStringToArray($string, $encoding = 'UTF-8') { 
         $arrayResult = 배열(); 
         while ($iLen = mb_strlen($string, $encoding)) { 
             array_push($arrayResult, mb_substr($string, 0, 1, $encoding)); 
             $string = mb_substr($string, 1, $iLen, $encoding); 
         } 
         $arrayResult를 반환합니다. 
     } 
     //编辑距离 
     function levenshtein_cn($str1, $str2, $costReplace = 1, $encoding = 'UTF-8') { 
         $count_same_letter = 0; 
         $d = 배열(); 
         $mb_len1 = mb_strlen($str1, $encoding); 
         $mb_len2 = mb_strlen($str2, $encoding); 
         $mb_str1 = mbStringToArray($str1, $encoding); 
         $mb_str2 = mbStringToArray($str2, $encoding); 
         for ($i1 = 0; $i1              $d[$i1] = 배열(); 
             $d[$i1][0] = $i1; 
         } 
         for ($i2 = 0; $i2              $d[0][$i2] = $i2; 
         } 
         for ($i1 = 1; $i1              for ($i2 = 1; $i2                  // $cost = ($str1[$i1 - 1] == $str2[$i2 - 1]) ? 0:1; 
                 if ($mb_str1[$i1 - 1] === $mb_str2[$i2 - 1]) { 
                     $ 비용 = 0; 
                     $count_same_letter ; 
                 } else { 
                     $cost = $costReplace; //替换 
                 } 
                 $d[$i1][$i2] = min($d[$i1 - 1][$i2] 1, //插入 
                 $d[$i1][$i2 - 1] 1, //删除 
                 $d[$i1 - 1][$i2 - 1] $비용); 
             } 
         } 
         $d[$mb_len1][$mb_len2] 반환; 
         //return array('distance' => $d[$mb_len1][$mb_len2], 'count_same_letter' => $count_same_letter); 
     }  

 
最长公共子序列LCS()
 

复代码 代码如下:

                  //最长公共子序列英文版
         함수 LCS_en($str_1, $str_2) { 
           $len_1 = strlen($str_1); 
           $len_2 = strlen($str_2); 
           $len = $len_1 > $len_2 ? $len_1 : $len_2; 
           $dp = 배열(); 
           for ($i = 0; $i              $dp[$i] = 배열(); 
             $dp[$i][0] = 0; 
             $dp[0][$i] = 0; 
           } 
           for ($i = 1; $i              for ($j = 1; $j                if ($str_1[$i - 1] == $str_2[$j - 1]) { 
                 $dp[$i][$j] = $dp[$i - 1][$j - 1] 1; 
               } else { 
                 $dp[$i][$j] = $dp[$i - 1][$j] > $dp[$i][$j - 1] ? $dp[$i - 1][$j] : $dp[$i][$j - 1]; 
               } 
             } 
           } 
           $dp[$len_1][$len_2] 반환; 
         } 
         //拆分字符串 
         function mbStringToArray($string, $encoding = 'UTF-8') { 
           $arrayResult = 배열(); 
           while ($iLen = mb_strlen($string, $encoding)) { 
             array_push($arrayResult, mb_substr($string, 0, 1, $encoding)); 
             $string = mb_substr($string, 1, $iLen, $encoding); 
           } 
           $arrayResult를 반환합니다. 
         } 
         //最长公共子序列中文版
         함수 LCS_cn($str1, $str2, $encoding = 'UTF-8') { 
           $mb_len1 = mb_strlen($str1, $encoding); 
           $mb_len2 = mb_strlen($str2, $encoding); 
           $mb_str1 = mbStringToArray($str1, $encoding); 
           $mb_str2 = mbStringToArray($str2, $encoding); 
           $len = $mb_len1 > $mb_len2 ? $mb_len1 : $mb_len2; 
           $dp = 배열(); 
           for ($i = 0; $i              $dp[$i] = 배열(); 
             $dp[$i][0] = 0; 
             $dp[0][$i] = 0; 
           } 
           for ($i = 1; $i              for ($j = 1; $j                if ($mb_str1[$i - 1] == $mb_str2[$j - 1]) { 
                 $dp[$i][$j] = $dp[$i - 1][$j - 1] 1; 
               } else { 
                 $dp[$i][$j] = $dp[$i - 1][$j] > $dp[$i][$j - 1] ? $dp[$i - 1][$j] : $dp[$i][$j - 1]; 
               } 
             } 
           } 
           $dp[$mb_len1][$mb_len2] 반환; 
         }