PHP从一个文本字符串中提取关键字的函数
这是一个函数定位接收一个字符串作为参数（连同其他配置可选参数），并且定位该字符串中的所有关键字（出现最多的词），返回一个数组或一个字符串由逗号分隔的关键字。
01 |
/** |
02 |
* Finds all of the keywords (words that appear most) on param $str |
03 |
* and return them in order of most occurrences to less occurrences. |
04 |
* @param string $str The string to search for the keywords. |
05 |
* @param int $minWordLen[optional] The minimun length (number of chars) of a word to be considered a keyword. http://www.zzzyk.com/ |
06 |
* @param int $minWordOccurrences[optional] The minimun number of times a word has to appear |
07 |
* on param $str to be considered a keyword. |
08 |
* @param boolean $asArray[optional] Specifies if the function returns a string with the |
09 |
* keywords separated by a comma ($asArray = false) or a keywords array ($asArray = true). |
10 |
* @return mixed A string with keywords separated with commas if param $asArray is true, |
11 |
* an array with the keywords otherwise. |
12 |
*/ |
13 |
function extract_keywords($str, $minWordLen = 3, $minWordOccurrences = 2, $asArray = false) |
14 |
{ |
15 |
function keyword_count_sort($first, $sec) |
16 |
{ |
17 |
return $sec[1] - $first[1]; |
18 |
} |
19 |
$str = preg_replace('/[^\\w0-9 ]/', ' ', $str); |
20 |
$str = trim(preg_replace('/\s+/', ' ', $str)); |
21 |
|
22 |
$words = explode(' ', $str); |
23 |
$keywords = array(); |
24 |
while(($c_word = array_shift($words)) !== null) |
25 |
{ |
26 |
if(strlen($c_word) <= $minWordLen) continue; |
27 |
|
28 |
$c_word = strtolower($c_word); |
29 |
if(array_key_exists($c_word, $keywords)) $keywords[$c_word][1]++; |
30 |
else $keywords[$c_word] = array($c_word, 1); |
31 |
} |
32 |
usort($keywords, 'keyword_count_sort'); |
33 |
|
34 |
$final_keywords = array(); |
35 |
foreach($keywords as $keyword_det) |
36 |
{ |
37 |
if($keyword_det[1] < $minWordOccurrences) break; |
38 |
array_push($final_keywords, $keyword_det[0]); |
39 |
} |
40 |
return $asArray ? $final_keywords : implode(', ', $final_keywords); |
41 |
} |
42 |
|
43 |
//How to use |
44 |
|
45 |
//Basic lorem ipsum text to extract the keywords |
46 |
$text = " |
47 |
Lorem ipsum dolor sit amet, consectetur adipiscing elit. |
48 |
Curabitur eget ipsum ut lorem laoreet porta a non libero. |
49 |
Vivamus in tortor metus. Suspendisse potenti. Curabitur |
50 |
metus nisi, adipiscing eget placerat suscipit, suscipit |
51 |
vitae felis. Integer eu odio enim, sed dignissim lorem. |
52 |
In fringilla molestie justo, vitae varius risus lacinia ac. |
53 |
Nulla port易做图 justo a lectus iaculis ut vestibulum magna |
54 |
egestas. Ut sed purus et nibh cursus fringilla at id purus. |
55 |
"; |
56 |
//Echoes: lorem, suscipit, metus, fringilla, purus, justo, eget, vitae, ipsum, curabitur, adipiscing |
57 |
echo extract_keywords($text); |
58 |
|
