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

similar_text()中文汉字版

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

levenshtein()中文汉字版

<br />     <?php  <br />     //拆分字符串  <br />     function mbStringToArray($string, $encoding = 'UTF-8') {  <br />         $arrayResult = array();  <br />         while ($iLen = mb_strlen($string, $encoding)) {  <br />             array_push($arrayResult, mb_substr($string, 0, 1, $encoding));  <br />             $string = mb_substr($string, 1, $iLen, $encoding);  <br />         }  <br />         return $arrayResult;  <br />     }  <br />     //编辑距离  <br />     function levenshtein_cn($str1, $str2, $costReplace = 1, $encoding = 'UTF-8') {  <br />         $count_same_letter = 0;  <br />         $d = array();  <br />         $mb_len1 = mb_strlen($str1, $encoding);  <br />         $mb_len2 = mb_strlen($str2, $encoding);  <br />         $mb_str1 = mbStringToArray($str1, $encoding);  <br />         $mb_str2 = mbStringToArray($str2, $encoding);  <br />         for ($i1 = 0; $i1 <= $mb_len1; $i1++) {  <br />             $d[$i1] = array();  <br />             $d[$i1][0] = $i1;  <br />         }  <br />         for ($i2 = 0; $i2 <= $mb_len2; $i2++) {  <br />             $d[0][$i2] = $i2;  <br />         }  <br />         for ($i1 = 1; $i1 <= $mb_len1; $i1++) {  <br />             for ($i2 = 1; $i2 <= $mb_len2; $i2++) {  <br />                 // $cost = ($str1[$i1 - 1] == $str2[$i2 - 1]) ? 0 : 1;  <br />              <mark>@本文来源gaodaimacom搞#代%码@网-</mark><strong>搞代gaodaima码</strong>   if ($mb_str1[$i1 - 1] === $mb_str2[$i2 - 1]) {  <br />                     $cost = 0;  <br />                     $count_same_letter++;  <br />                 } else {  <br />                     $cost = $costReplace; //替换  <br />                 }  <br />                 $d[$i1][$i2] = min($d[$i1 - 1][$i2] + 1, //插入  <br />                 $d[$i1][$i2 - 1] + 1, //删除  <br />                 $d[$i1 - 1][$i2 - 1] + $cost);  <br />             }  <br />         }  <br />         return $d[$mb_len1][$mb_len2];  <br />         //return array('distance' => $d[$mb_len1][$mb_len2], 'count_same_letter' => $count_same_letter);  <br />     }   <br />

最长公共子序列LCS()

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

喜欢 (0)赏

[搞代码]

分享 (0)