163 lines
6.6 KiB
Markdown
163 lines
6.6 KiB
Markdown
# pycorrector
|
||
[](https://github.com/deepmipt/DeepPavlov/blob/master/LICENSE)  
|
||
|
||
中文错别字纠正工具。音似、形似错字（或变体字）纠正，可用于中文拼音、笔画输入法的错误纠正。python3开发。
|
||
|
||
**pycorrector**依据语言模型检测错别字位置，通过拼音音似特征、笔画五笔编辑距离特征及语言模型困惑度特征纠正错别字。
|
||
|
||
## 特征
|
||
### 语言模型
|
||
* Kenlm（统计语言模型工具）
|
||
* RNNLM（TensorFlow、PaddlePaddle均有实现栈式双向LSTM的语言模型）
|
||
|
||
### 错误检测
|
||
* 字粒度：语言模型困惑度（ppl）检测某字的似然概率值低于句子文本平均值，则判定该字是疑似错别字的概率大。
|
||
* 词粒度：切词后不在词典中的词是疑似错词的概率大。
|
||
|
||
|
||
### 错误纠正
|
||
1. 通过错误检测定位所有疑似错误后，取所有疑似错字的音似、形似候选词，
|
||
2. 使用候选词替换，基于语言模型得到类似翻译模型的候选排序结果，得到最优纠正词。
|
||
|
||
|
||
### 思考
|
||
1. 现在的处理手段，在词粒度的错误召回还不错，但错误纠正的准确率还有待提高，更多优质的纠错集及纠错词库会有提升，我更希望算法上有更大的突破。
|
||
2. 另外，现在的文本错误不再局限于字词粒度上的拼写错误，需要提高中文语法错误检测（CGED, Chinese Grammar Error Diagnosis）及纠正能力，列在TODO中，后续调研。
|
||
|
||
## demo
|
||
|
||
http://www.borntowin.cn/nlp/corrector.html
|
||
|
||
|
||
## 使用说明
|
||
|
||
### 依赖
|
||
pip3 install -r requirements.txt
|
||
|
||
pip3 install git+https://www.github.com/keras-team/keras-contrib.git
|
||
|
||
|
||
### 安装
|
||
* 全自动安装：pip3 install pycorrector
|
||
* 半自动安装：
|
||
```
|
||
git clone https://github.com/shibing624/pycorrector.git
|
||
cd pycorrector
|
||
python3 setup.py install
|
||
```
|
||
|
||
|
||
### 纠错
|
||
使用示例:
|
||
```
|
||
import pycorrector
|
||
|
||
corrected_sent, detail = pycorrector.correct('少先队员因该为老人让坐')
|
||
print(corrected_sent, detail)
|
||
|
||
```
|
||
|
||
输出:
|
||
```
|
||
少先队员应该为老人让座 [[('因该', '应该', 4, 6)], [('坐', '座', 10, 11)]]
|
||
```
|
||
|
||
## 自定义语言模型
|
||
|
||
语言模型对于纠错步骤至关重要，目前我能收集到的语料数据有人民日报数据。大家可以用中文维基（繁体转简体，pycorrector.utils下有此功能）等更大的语料数据训练效果更好的语言模型，
|
||
对于纠错效果会有比较好的提升。
|
||
|
||
1. kenlm语言模型训练工具的使用，请见博客：http://blog.csdn.net/mingzai624/article/details/79560063
|
||
2. 附上训练语料<人民日报2014版熟语料>，包括：
|
||
1）标准人工切词及词性数据people2014.tar.gz，
|
||
2）未切词文本数据people2014_words.txt，
|
||
3）kenlm训练字粒度语言模型文件及其二进制文件people2014corpus_chars.arps/klm，
|
||
4）kenlm词粒度语言模型文件及其二进制文件people2014corpus_words.arps/klm。
|
||
|
||
网盘链接:https://pan.baidu.com/s/1971a5XLQsIpL0zL0zxuK2A 密码:uc11。尊重版权，传播请注明出处。
|
||
|
||
## 贡献及优化点（TODO）
|
||
|
||
* 使用RNN语言模型来提高纠错准确率。
|
||
* 优化形似字字典，提高形似字纠错准确率。
|
||
* 整理中文纠错集，使用seq2seq做深度中文纠错模型。
|
||
* 添加中文语法错误检测及纠正能力。
|
||
|
||
## 参考
|
||
|
||
1. [基于文法模型的中文纠错系统](https://blog.csdn.net/mingzai624/article/details/82390382)
|
||
2. [Norvig’s spelling corrector](http://norvig.com/spell-correct.html)
|
||
3. [《Chinese Spelling Error Detection and Correction Based on Language Model, Pronunciation, and Shape》[Yu, 2013]](http://www.aclweb.org/anthology/W/W14/W14-6835.pdf)
|
||
4. [《Chinese Spelling Checker Based on Statistical Machine Translation》[Chiu, 2013]](http://www.aclweb.org/anthology/O/O13/O13-1005.pdf)
|
||
5. [《Chinese Word Spelling Correction Based on Rule Induction》[yeh, 2014]](http://aclweb.org/anthology/W14-6822)
|
||
6. [《Neural Language Correction with Character-Based Attention》[Ziang Xie, 2016]](https://arxiv.org/pdf/1603.09727.pdf)
|
||
|
||
|
||
----
|
||
|
||
# pycorrector
|
||
Chinese text error correction tool.
|
||
|
||
|
||
**pycorrector** Use the language model to detect errors, pinyin feature and shape feature to correct chinese text
|
||
error, it can be used for Chinese Pinyin and stroke input method.
|
||
|
||
## Features
|
||
### language model
|
||
* Kenlm
|
||
* RNNLM
|
||
|
||
## Usage
|
||
|
||
### install
|
||
* pip install pycorrector / pip3 install pycorrector
|
||
* Or download https://github.com/shibing624/pycorrector, Unzip and run: python setup.py install
|
||
|
||
### correct
|
||
input:
|
||
```
|
||
import pycorrector
|
||
|
||
corrected_sent, detail = pycorrector.correct('少先队员因该为老人让坐')
|
||
print(corrected_sent, detail)
|
||
|
||
```
|
||
|
||
output:
|
||
```
|
||
少先队员应该为老人让座 [[('因该', '应该', 4, 6)], [('坐', '座', 10, 11)]]
|
||
```
|
||
|
||
|
||
### Future work
|
||
1. P(c), the language model. We could create a better language model by collecting more data, and perhaps by using a
|
||
little English morphology (such as adding "ility" or "able" to the end of a word).
|
||
|
||
2. P(w|c), the error model. So far, the error model has been trivial: the smaller the edit distance, the smaller the
|
||
error.
|
||
Clearly we could use a better model of the cost of edits. get a corpus of spelling errors, and count how likely it is
|
||
to make each insertion, deletion, or alteration, given the surrounding characters.
|
||
|
||
3. It turns out that in many cases it is difficult to make a decision based only on a single word. This is most
|
||
obvious when there is a word that appears in the dictionary, but the test set says it should be corrected to another
|
||
word anyway:
|
||
correction('where') => 'where' (123); expected 'were' (452)
|
||
We can't possibly know that correction('where') should be 'were' in at least one case, but should remain 'where' in
|
||
other cases. But if the query had been correction('They where going') then it seems likely that "where" should be
|
||
corrected to "were".
|
||
|
||
4. Finally, we could improve the implementation by making it much faster, without changing the results. We could
|
||
re-implement in a compiled language rather than an interpreted one. We could cache the results of computations so
|
||
that we don't have to repeat them multiple times.
|
||
One word of advice: before attempting any speed optimizations, profile carefully to see where the time is actually
|
||
going.
|
||
|
||
|
||
### Further Reading
|
||
* [Roger Mitton has a survey article on spell checking.](http://www.dcs.bbk.ac.uk/~roger/spellchecking.html)
|
||
|
||
# Reference
|
||
1. [Norvig’s spelling corrector](http://norvig.com/spell-correct.html)
|
||
2. [Norvig’s spelling corrector(java version)](http://raelcunha.com/spell-correct/)
|
||
|