Skip to content

一款高性能敏感词(非法词/脏字)检测过滤组件,附带繁体简体互换,支持全角半角互换,汉字转拼音,模糊搜索等功能。

License

Notifications You must be signed in to change notification settings

toolgood/ToolGood.Words

Folders and files

NameName
Last commit message
Last commit date

Latest commit

author
linzhijun
Sep 26, 2024
c37a350 · Sep 26, 2024
Sep 26, 2024
Sep 25, 2024
Aug 31, 2024
Oct 23, 2023
May 17, 2020
May 23, 2020
Feb 23, 2024
Oct 23, 2023
Jan 18, 2020
Jan 13, 2017
Apr 11, 2020
Apr 11, 2020
Apr 11, 2020
Apr 11, 2020
Apr 11, 2020
May 24, 2020
Sep 1, 2020
Dec 19, 2016
Jan 13, 2024

Repository files navigation

ToolGood.Words

一款高性能非法词(敏感词)检测组件,附带繁体简体互换,支持全角半角互换,获取拼音首字母,获取拼音字母,拼音模糊搜索等功能。

C#语言,使用StringSearchEx2.Replace过滤,在48k敏感词库上的过滤速度超过3亿字符每秒。(cpu i7 8750h)

csharp 文件夹说明:

ToolGood.Pinyin.Build:          生成词的拼音
ToolGood.Pinyin.Pretreatment:   生成拼音预处理,核对拼音,词组最小化
ToolGood.Transformation.Build: 生成简体繁体转换文档,更新时文档放在同一目录下,词库参考 https://github.com/BYVoid/OpenCC
ToolGood.Words.Contrast:        字符串搜索对比
ToolGood.Words.Test:            单元测试
ToolGood.Words:                 本项目源代码

非法词(敏感词)检测(字符串搜索)

非法词(敏感词)检测类:StringSearchStringSearchExStringSearchEx2WordsSearchWordsSearchExWordsSearchEx2IllegalWordsSearch;

  • StringSearchStringSearchExStringSearchEx2StringSearchEx3: 搜索FindFirst方法返回结果为string类型。
  • WordsSearchWordsSearchExWordsSearchEx2WordsSearchEx3: 搜索FindFirst方法返回结果为WordsSearchResult类型, WordsSearchResult不仅仅有关键字,还有关键字的开始位置、结束位置,关键字序号等。
  • IllegalWordsSearch: 过滤非法词(敏感词)专用类,可设置跳字长度,默认全角转半角,忽略大小写,跳词,重复词,黑名单, 搜索FindFirst方法返回为IllegalWordsSearchResult,有关键字,对应原文,开始、位置,黑名单类型。
  • IllegalWordsSearchStringSearchExStringSearchEx2WordsSearchExWordsSearchEx2 使用SaveLoad方法,可以加快初始化。
  • 共同方法有:SetKeywordsContainsAnyFindFirstFindAllReplace
  • IllegalWordsSearch独有方法:SetSkipWords(设置跳词)、SetBlacklist(设置黑名单)。
  • IllegalWordsSearch字段UseIgnoreCase:设置是忽略否大小写,必须在SetKeywords方法之前,注:使用Load方法则该字段无效。
  • StringSearchEx3WordsSearchEx3为指针版优化版,实测时发现性能浮动比较大。
    string s = "中国|国人|zg人";
    string test = "我是中国人";

    StringSearch iwords = new StringSearch();
    iwords.SetKeywords(s.Split('|'));
    
    var b = iwords.ContainsAny(test);
    Assert.AreEqual(true, b);

    var f = iwords.FindFirst(test);
    Assert.AreEqual("中国", f);

    var all = iwords.FindAll(test);
    Assert.AreEqual("中国", all[0]);
    Assert.AreEqual("国人", all[1]);
    Assert.AreEqual(2, all.Count);

    var str = iwords.Replace(test, '*');
    Assert.AreEqual("我是***", str);

非法词(敏感词)检测(字符串搜索)(支持通配符)

非法词(敏感词)检测类:StringMatchStringMatchExWordsMatchWordsMatchEx

支持部分正则表达式类型:.(点)?(问号) [](方括号) (|)(括号与竖线)

    string s = ".[中美]国|国人|zg人";
    string test = "我是中国人";

    WordsMatch wordsSearch = new WordsMatch();
    wordsSearch.SetKeywords(s.Split('|'));

    var b = wordsSearch.ContainsAny(test);
    Assert.AreEqual(true, b);

    var f = wordsSearch.FindFirst(test);
    Assert.AreEqual("是中国", f.Keyword);

    var alls = wordsSearch.FindAll(test);
    Assert.AreEqual("是中国", alls[0].Keyword);
    Assert.AreEqual(".[中美]国", alls[0].MatchKeyword);
    Assert.AreEqual(1, alls[0].Start);
    Assert.AreEqual(3, alls[0].End);
    Assert.AreEqual(0, alls[0].Index);//返回索引Index,默认从0开始
    Assert.AreEqual("国人", alls[1].Keyword);
    Assert.AreEqual(2, alls.Count);

    var t = wordsSearch.Replace(test, '*');
    Assert.AreEqual("我****", t);

繁体简体互换、全角半角互换、数字转成中文大写、拼音操作

    // 转成简体
    WordsHelper.ToSimplifiedChinese("我愛中國");
    WordsHelper.ToSimplifiedChinese("我愛中國",1);// 港澳繁体 转 简体
    WordsHelper.ToSimplifiedChinese("我愛中國",2);// 台湾正体 转 简体
    // 转成繁体
    WordsHelper.ToTraditionalChinese("我爱中国");
    WordsHelper.ToTraditionalChinese("我爱中国",1);// 简体 转 港澳繁体
    WordsHelper.ToTraditionalChinese("我爱中国",2);// 简体 转 台湾正体
    // 转成全角
    WordsHelper.ToSBC("abcABC123");
    // 转成半角
    WordsHelper.ToDBC("abcABC123");
    // 数字转成中文大写
    WordsHelper.ToChineseRMB(12345678901.12);
    // 中文转成数字
    WordsHelper.ToNumber("壹佰贰拾叁亿肆仟伍佰陆拾柒万捌仟玖佰零壹元壹角贰分");
    // 获取全拼
    WordsHelper.GetPinyin("我爱中国");//WoAiZhongGuo   
    WordsHelper.GetPinyin("我爱中国",",");//Wo,Ai,Zhong,Guo   
    WordsHelper.GetPinyin("我爱中国",true);//WǒÀiZhōngGuó

    // 获取首字母
    WordsHelper.GetFirstPinyin("我爱中国");//WAZG
    // 获取全部拼音
    WordsHelper.GetAllPinyin('传');//Chuan,Zhuan
    // 获取姓名
    WordsHelper.GetPinyinForName("单一一")//ShanYiYi
    WordsHelper.GetPinyinForName("单一一",",")//Shan,Yi,Yi
    WordsHelper.GetPinyinForName("单一一",true)//ShànYīYī
拼音分支

ToolGood.Words.Pinyin 追求更快的加载速度(目前只有C#代码)。

拼音匹配

PinyinMatch:方法有SetKeywordsSetIndexsFindFindIndex

PinyinMatch<T>:方法有SetKeywordsFuncSetPinyinFuncSetPinyinSplitCharFind

    string s = "北京|天津|河北|辽宁|吉林|黑龙江|山东|江苏|上海|浙江|安徽|福建|江西|广东|广西|海南|河南|湖南|湖北|山西|内蒙古|宁夏|青海|陕西|甘肃|新疆|四川|贵州|云南|重庆|西藏|香港|澳门|台湾";

    PinyinMatch match = new PinyinMatch();
    match.SetKeywords(s.Split('|').ToList());

    var all = match.Find("BJ");
    Assert.AreEqual("北京", all[0]);
    Assert.AreEqual(1, all.Count);

    all = match.Find("北J");
    Assert.AreEqual("北京", all[0]);
    Assert.AreEqual(1, all.Count);

    all = match.Find("北Ji");
    Assert.AreEqual("北京", all[0]);
    Assert.AreEqual(1, all.Count);

    all = match.Find("S");
    Assert.AreEqual("山东", all[0]);
    Assert.AreEqual("江苏", all[1]);

    var all2 = match.FindIndex("BJ");
    Assert.AreEqual(0, all2[0]);
    Assert.AreEqual(1, all.Count);

性能对比

执行10万次性能对比,结果如下:

10W次性能对比

注:C#自带正则很慢,StringSearchEx2.ContainsAnyRegex.IsMatch效率的8.8万倍多,跟关键字数量有关。

Regex.Matches的运行方式跟IQueryable的类似,只返回MatchCollection,还没有计算。

在 Find All测试中, (检测出的文本中有敏感词汇,就不显示了,大家可自行调试查看)。

FastFilter只能检测出7个

StringSearch检测出14个

插曲:在细查Regex.Matches神奇3ms,我发现Regex.Matches有一个小问题,

Regex.Matches只能检测出11个

其他语言实现

Lua版本

作者:wenlifan 地址:https://github.com/wenlifan/SensitiveWordFilter

推荐一下

《ToolGood 内容审核系统》正式开源,Windows、Linux双平台,内存占用小于100M。

官网:https://toolgood.com/

开源代码:https://github.com/toolgood/ToolGood.TextFilter

敏感信息过滤研究会,Q群:128994346(已满)

本人不是老师,请不要提关于项目使用、加载等简单问题。

敏感词相关文章

1、敏感词过滤方案那些事

2、普通公司敏感词审核制度

3、新人小白过滤敏感词方案

4、网络常用敏感词过滤方法

5、ToolGood.Words算法过滤敏感词优化原理 (收费30元,一顿KFC)

6、ToolGood.TextFilter开源代码优化详解 (收费300元) 与IllegalWordsSearch算法进行对比,阐述了ToolGood.TextFilter过滤算法优化点,如何减少内存使用量。 还有一小部分未写好,心急的人可以先买,我会持续更新。

7、正则转DFA算法(C#版、JAVA版) (收费30元,一顿KFC) ToolGood.TextFilter的一个核心算法就使用到正则转DFA。

8、C#版图片鉴黄(收费30元,一顿KFC)

比特币私钥碰撞机

比特币私钥碰撞机,利用电脑空闲性能(3G内存),搏25万枚比特币。

比特币私钥碰撞机(收费50元)

比特币私钥碰撞机源码(收费500元)