一個用PHP寫的中文分詞函數

來源:互聯網
上載者:User

 

<?php

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;?>



相關文章

Beyond APAC's No.1 Cloud

19.6% IaaS Market Share in Asia Pacific - Gartner IT Service report, 2018

Learn more >

Apsara Conference 2019

The Rise of Data Intelligence, September 25th - 27th, Hangzhou, China

Learn more >

Alibaba Cloud Free Trial

Learn and experience the power of Alibaba Cloud with a free trial worth $300-1200 USD

Learn more >

聯繫我們

該頁面正文內容均來源於網絡整理,並不代表阿里雲官方的觀點,該頁面所提到的產品和服務也與阿里云無關,如果該頁面內容對您造成了困擾,歡迎寫郵件給我們,收到郵件我們將在5個工作日內處理。

如果您發現本社區中有涉嫌抄襲的內容,歡迎發送郵件至: info-contact@alibabacloud.com 進行舉報並提供相關證據,工作人員會在 5 個工作天內聯絡您,一經查實,本站將立刻刪除涉嫌侵權內容。