public static function segments($arr, $num = 10)
{
$list = array();
if (empty($text)) {
return $list;
}
$words = array();
//检测是否已安装php_scws扩展
if (function_exists("scws_open")) {
$sh = scws_open();
scws_set_charset($sh, 'utf8');
scws_set_dict($sh, APP_ROOT_PATH . 'system/scws/dict.utf8.xdb');
scws_set_rule($sh, APP_ROOT_PATH . 'system/scws/rules.utf8.ini');
scws_set_ignore($sh, true);
foreach ($arr as $key => $text) {
scws_send_text($sh, $text);
$words[] = scws_get_tops($sh, $num);
}
scws_close($sh);
} else {
require_once APP_ROOT_PATH . 'system/scws/pscws4.class.php';
$pscws = new PSCWS4();
$pscws->set_dict(APP_ROOT_PATH . 'system/scws/dict.utf8.xdb');
$pscws->set_rule(APP_ROOT_PATH . 'system/scws/rules.utf8.ini');
$pscws->set_ignore(true);
foreach ($arr as $key => $text) {
$pscws->send_text($text);
$words[] = $pscws->get_tops($num);
}
$pscws->close();
}
for ($i = 0; $i < $num; $i++) {
foreach ($words as $item) {
if (isset($item[$i])) {
$word = $item[$i]['word'];
if (isset($list[$word])) {
$list[$word]++;
} else {
$list[$word] = 1;
}
}
}
}
$list = array_slice($list, 0, $num);
return array_keys($list);
}
/**
* 文本分词
* @param string $text 需要分词的文本
* @return array
*/
public static function segmentAll($text)
{
$list = array();
if (empty($text)) {
return $list;
}
//检测是否已安装php_scws扩展
if (function_exists("scws_open")) {
$sh = scws_open();
scws_set_charset($sh, 'utf8');
scws_set_dict($sh, APP_ROOT_PATH . 'system/scws/dict.utf8.xdb');
scws_set_rule($sh, APP_ROOT_PATH . 'system/rules.utf8.ini');
scws_set_ignore($sh, true);
scws_send_text($sh, $text);
while ($words = scws_get_result($sh)) {
foreach ($words as $word) {
$list[] = $word['word'];
}
}
scws_close($sh);
} else {
require_once APP_ROOT_PATH . 'system/scws/pscws4.class.php';
$pscws = new PSCWS4();
$pscws->set_dict(APP_ROOT_PATH . 'system/scws/dict.utf8.xdb');
$pscws->set_rule(APP_ROOT_PATH . 'system/scws/rules.utf8.ini');
$pscws->set_ignore(true);
$pscws->send_text($text);
while ($words = $pscws->get_result()) {
foreach ($words as $word) {
$list[] = $word['word'];
}
}
$pscws->close();
}
return $list;
}
scws_set_rule($cws, ini_get('scws.default.fpath') . '/rules.utf8.ini');
//scws_set_ignore($cws, true);
//scws_set_multi($cws, true);
scws_send_text($cws, $text);
echo "\n";
// top words
printf("No. WordString Attr Weight(times)\n");
printf("-------------------------------------------------\n");
$list = scws_get_tops($cws, 10, "~v");
$cnt = 1;
foreach ($list as $tmp) {
printf("%02d. %-24.24s %-4.2s %.2f(%d)\n", $cnt, $tmp['word'], $tmp['attr'], $tmp['weight'], $tmp['times']);
$cnt++;
}
echo "\n\n-------------------------------------------------\n";
// segment
while ($res = scws_get_result($cws)) {
foreach ($res as $tmp) {
if ($tmp['len'] == 1 && $tmp['word'] == "\r") {
continue;
}
if ($tmp['len'] == 1 && $tmp['word'] == "\n") {
echo $tmp['word'];
} else {
printf("%s/%s ", $tmp['word'], $tmp['attr']);
}
}
}
echo "\n\n";
scws_close($cws);
/**
* 分词处理，取出词频最高的词组，并可以指定词性进行查找
*
* @param string $str
* @param int $limit
* 可选参数，返回的词的最大数量，缺省是 10
* @param string $attr
* 可选参数，是一系列词性组成的字符串，各词性之间以半角的逗号隔开， 这表示返回的词性必须在列表中，如果以~开头，则表示取反，词性必须不在列表中，缺省为NULL，返回全部词性，不过滤。
* @return multitype:
*/
function scwsTop($str, $limit = 10, $attr = null)
{
if (!function_exists('scws_open')) {
return false;
}
$rst = array();
$str = preg_replace("/[\\s\t\r\n]+/", '', $str);
if (!empty($str)) {
$sh = scws_open();
scws_set_charset($sh, 'utf8');
scws_set_ignore($sh, true);
scws_set_multi($sh, SCWS_MULTI_SHORT | SCWS_MULTI_DUALITY);
scws_set_duality($sh, true);
scws_send_text($sh, $str);
$rst = scws_get_tops($sh, $limit, $attr);
scws_close($sh);
}
return $rst;
}
public function close()
{
scws_close($this->tool);
}