Git Product home page Git Product logo

Comments (8)

f1120309265 avatar f1120309265 commented on July 29, 2024 1

okay,谢谢解答。
我需要 调整权重的词比较多,所以继承MyJiebaAnalyse类,添加了一个方法,自己载入idf文件,与默认的合并,这样就解决了权重的问题。

from jieba-php.

fukuball avatar fukuball commented on July 29, 2024

@f1120309265 你可以提供一下這個例子的 $content 嗎?

from jieba-php.

fukuball avatar fukuball commented on July 29, 2024

@f1120309265 我仔細看一下 jieba 目前的實作邏輯,自定義詞庫的權重影響的是斷詞切分的組合,跟取出關鍵字的 tf/idf 並無關係,如果你有調整關鍵字權重的需求,我可能要另外實作方法,而且也不是用自定義詞庫來調,可能就提供發法在 run time 中去調整 tf/idf 值。

from jieba-php.

fukuball avatar fukuball commented on July 29, 2024

剛剛仔細看源碼,應該不用實作新方法就可以達到您的需求,提供一個範例在下面。

from jieba-php.

fukuball avatar fukuball commented on July 29, 2024

@f1120309265 提供一個範例給你:

$content = "汽车行业巨头大众汽车最新公布2018年销售数据,全年共销售汽车624万辆,创历史新高。尽管如此,总体销量较2017年只是小幅增长0.2%,尤其是下半年受贸易问题和欧洲排放标准的影响,销量增速快速恶化。";

$tags = JiebaAnalyse::extractTags($content, 10);
var_dump($tags);

Jieba::loadUserDict(dirname(dirname(__FILE__)).'/dict/user_dict.txt');

$tags = JiebaAnalyse::extractTags($content, 10);
var_dump($tags);

JiebaAnalyse::$idf_freq['销售数据'] = 20;
$tags = JiebaAnalyse::extractTags($content, 10);
var_dump($tags);

Output 的結果:

array(10) {
  ["销量"]=>
  float(0.38517762539314)
  [624]=>
  float(0.34156478579714)
  ["0.2"]=>
  float(0.34156478579714)
  [2017]=>
  float(0.34156478579714)
  [2018]=>
  float(0.34156478579714)
  ["销售"]=>
  float(0.30390049575086)
  ["大众汽车"]=>
  float(0.284269540954)
  ["汽车行业"]=>
  float(0.24650508838314)
  ["尽管如此"]=>
  float(0.23087413490343)
  ["巨头"]=>
  float(0.21458391385457)
}
array(10) {
  ["销量"]=>
  float(0.39650637908118)
  [624]=>
  float(0.35161080890882)
  ["0.2"]=>
  float(0.35161080890882)
  [2017]=>
  float(0.35161080890882)
  [2018]=>
  float(0.35161080890882)
  ["销售数据"]=>
  float(0.35161080890882)
  ["大众汽车"]=>
  float(0.29263040980559)
  ["汽车行业"]=>
  float(0.25375523804147)
  ["尽管如此"]=>
  float(0.23766455063588)
  ["巨头"]=>
  float(0.22089520543853)
}
array(10) {
  ["销售数据"]=>
  float(0.58823529411765)
  ["销量"]=>
  float(0.39650637908118)
  [624]=>
  float(0.35161080890882)
  ["0.2"]=>
  float(0.35161080890882)
  [2017]=>
  float(0.35161080890882)
  [2018]=>
  float(0.35161080890882)
  ["大众汽车"]=>
  float(0.29263040980559)
  ["汽车行业"]=>
  float(0.25375523804147)
  ["尽管如此"]=>
  float(0.23766455063588)
  ["巨头"]=>
  float(0.22089520543853)
}

from jieba-php.

fukuball avatar fukuball commented on July 29, 2024

@f1120309265 從上面的範例,我們可以看到加了自定義字典之後,Jieba 可以斷出想要的結果,但是關鍵字的分數則不受影響(自定義字典中的權重意義上只是為了改善斷詞結果,而不是更改關鍵字演算結果),因為在正常情況下,我們不應該任意調整關鍵字演算法的權重計算結果。

但當有特別想要調整的關鍵字權重,我們還是可以使用 JiebaAnalyse::$idf_freq['關鍵字'] = 權重; 例如 JiebaAnalyse::$idf_freq['销售数据'] = 20; 來調整關鍵字權重結果,來對特定字詞調整 idf 值會比較符合你的使用情境,在意義上也會比較對。

from jieba-php.

sj8354661 avatar sj8354661 commented on July 29, 2024

okay,谢谢解答。
我需要 调整权重的词比较多,所以继承MyJiebaAnalyse类,添加了一个方法,自己载入idf文件,与默认的合并,这样就解决了权重的问题。

请问您写的这个方法可以分享一下么?我也有同样的需求!~

from jieba-php.

f1120309265 avatar f1120309265 commented on July 29, 2024

okay,谢谢解答。
我需要 调整权重的词比较多,所以继承MyJiebaAnalyse类,添加了一个方法,自己载入idf文件,与默认的合并,这样就解决了权重的问题。

请问您写的这个方法可以分享一下么?我也有同样的需求!~

代码如下,有问题再随时探讨沟通
`use Fukuball\Jieba\JiebaAnalyse;

class MyJiebaAnalyse extends JiebaAnalyse
{
public static function init($options = array())
{
parent::init($options = array());
}

public static function setIdf($idf_path)
{
    $content = fopen($idf_path, "r");
    while (($line = fgets($content)) !== false) {
        $explode_line = explode(" ", trim($line));
        $word = $explode_line[0];
        $freq = $explode_line[1];
        $freq = (float) $freq;
        $idffreq[$word] = $freq;
        self::$idf_freq[$word] = $freq;
    }
    fclose($content);
    asort(self::$idf_freq);
    $keys = array_keys(self::$idf_freq);
    $middle_key = $keys[count(self::$idf_freq)/2];
    self::$max_idf = max(self::$idf_freq);
    self::$median_idf = self::$idf_freq[$middle_key];
}

}`

from jieba-php.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.