This article mainly introduces the implementation of PHP in Chinese parts of speech, combined with a complete example of PHP based on string traversal, conversion, arithmetic and other techniques to achieve Chinese word segmentation function of the specific method, the need for friends can refer to the next
In this paper, we describe the Chinese parts of speech implemented by PHP. Share to everyone for your reference, as follows:
The Chinese division of Speech source code using Http://tools.jb51.net/code/jb51_php_format for the format of processing, easy to read. The specific code is as follows:
Class Segmentation {var $options = array (' lowercase ' = = TRUE, ' segment_english ' = FALSE); var $dict _name = ' Unknown '; var $dict _words = Array (); function Setlowercase ($value) {if ($value) {$this->options[' lowercase '] = TRUE; } else {$this->options[' lowercase ') = FALSE; } return TRUE; } function Setsegmentenglish ($value) {if ($value) {$this->options[' segment_english '] = TRUE; } else {$this->options[' segment_english ') = FALSE; } return TRUE; } function Load ($dict _file) {if (!file_exists ($dict _file)) {return FALSE; } $fp = fopen ($dict _file, ' R '); $temp = fgets ($fp, 1024); if ($temp = = = False) {return false; } else {if (Strpos ($temp, "\ T")!== FALSE) {list ($dict _type, $dict _name) = explode ("\ t", trim ($temp)); } else {$dict _type = trim ($temp); $dict _name = ' Unknown '; } $this->dict_name = $dict _name; if ($dict _type!== ' Dict_word_W ') {return FALSE; }} while (!feof ($fp)) {$this->dict_words[rtrim (fgets ($FP, 32))] = 1; } fclose ($FP); return TRUE; } function Getdictname () {return $this->dict_name; } function segmentstring ($STR) {if (count ($this->dict_words) = = = 0) {return FALSE; } $lines = explode ("\ n", $str); return $this->_segmentlines ($lines); } function Segmentfile ($filename) {if (count ($this->dict_words) = = = 0) {return FALSE; } $lines = file ($filename); return $this->_segmentlines ($lines); } function _segmentlines ($lines) {$contents _segmented = '; foreach ($lines as $line) {$contents _segmented. = $this->_segmentline (RTrim ($line)). "\ n"; } do {$contents _segmented = Str_replace (",", $contents _segmented); } while (Strpos ($contents _segmented, ")!== FALSE); return $contents _segmented; } function _segmentline ($str) {$str _final = '; $str _array = Array (); $Str_length = strlen ($STR); if ($str _length > 0) {if (ord ($str {$str _length-1}) >= 129) {$str. = '; }} for ($i =0; $i < $str _length; $i + +) {if (ord ($str {$i}) >= 129) {$str _array[] = $str {$i}. $str { $i +1}; $i + +; } else {$str _tmp = $str {$i}; for ($j = $i +1; $j < $str _length; $j + +) {if (ord ($str {$j}) < 129) {$str _tmp. = $str {$j}; } else {break; }} $str _array[] = Array ($str _tmp); $i = $j-1; }} $pos = count ($str _array); while ($pos > 0) {$char = $str _array[$pos-1]; if (Is_array ($char)) {$str _final_tmp = $char [0]; if ($this->options[' Segment_english ') {$str _final_tmp = Preg_replace ("/([\!\" \#\$\%\&\ ' \ (\) \*\+\,\-\.\/\ : \;\<\=\>\?\@\[\\\\\]\^\_\ ' \{\|\}\~\t\f]+)/"," $ ", $str _final_tmp); $str _final_tmp = Preg_replace ("/([\!\" \#\$\%\&\ ' \ (\) \*\+\,\-\.\/\:\;\<\=\>\?\@\[\\\\\]\^\_\ ' \{\|\}\~\t\f]) ([\!\ "\#\$\%\&\ ' \ (\) \*\+\,\-\.\/\:\;\<\=\>\?\@\[\\\\\]\^\_\ ' \{\|\}\~\t \f])/"," $ $ ", $str _final_tmp); } if ($this->options[' lowercase ') {$str _final_tmp = strtolower ($str _final_tmp); } $str _final = "$str _final_tmp$str_final"; $pos--; } else {$word _found = 0; $word _array = Array (0 = "); if ($pos < 4) {$word _temp = $pos + 1; } else {$word _temp = 5; } for ($i =1; $i < $word _temp; $i + +) {$word _array[$i] = $str _array[$pos-$i]. $word _array[$i-1]; } for ($i = ($word _temp-1), $i >1; $i-) {if (array_key_exists ($word _array[$i], $this->dict_words)) { $word _found = $i; Break }} if ($word _found) {$str _final = "$word _array[$word _found] $str _final"; $pos = $pos-$word _found; } else {$str _final = "$char $str_final"; $pos--; }}} return $str _final; }}?>