A framework for cleaning Chinese dialog data

Overview

本项目为一个清洗对话数据的多线程框架,针对知乎、微博、贴吧等。 目前还比较简陋,欢迎提bug和优化,比如句内重复短语降重函数的正则或者后缀算法。 代码还在继续完善中,注释以及一些函数出处引用等待完善。

目录结构

--scripts: 存放运行脚本
  ---run.sh: 使用我挑选的几个规则来运行run_dist.py  
--src: 清洗框架功能主目录  
  ---inputters: 存放dataloader 和 存取数据工具函数
  ---rules: 存放各级别的规则函数
  ---single_filter.py: run_dist.py所调用的单个线程的主程序,加载处理单个数据,并保存过滤后的数据以及脏   
---tool_data: 存放黑名单词典,每行一个词  
---run_dist.py: 主运行文件,加载dataloader,加载黑名单,简历线程池 
---utils: 数据统计,结果检测

运行并保存日志

bash ./scripts/run.sh 2>&1 | tee -a cleaning.log

Rules

规则包括目前大部分论文内的清洗规则:

1 黑名单过滤,包括特殊字符和脏话
2 emoji表情
3 邮箱、电话号等隐私过滤, 人名 替换为NAME1、NAME2。。。
4 URL过滤
5 unicode 相关修复
6 去重:包括重复词缩减、过滤掉上下文相同的句子、重复的对话
7 meena以及dialogpt中使用的广告、通用回复筛除

以上识别出来的噪音,如可在句中抹去则抹去。
如不可抹去则放弃该句子:即,若是单轮对话放弃该对话,若是多轮对话则以该句为分割,切分对话。

NOTE THAT: 1, 改动某规则的时候注意是否影响到其他规则, 规则清洗顺序有要求 2, 黑名单如人名、特殊话题等可根据需要配置放置到 ./tool_data/下,文件命名可自行配置请参阅。/run_dist.py中dataloader。黑名单可到github上搜寻,如 https://github.com/fighting41love/funNLP 3, 将在每个函数上方给定测试样例,下方给定期待样例 4, 目前run.sh中使用的参数为本人正在使用的功能

Auguments

参数 描述
n_p 多进程数
batch_size 单个进程最大处理session数
tool_dir 工具数据所在目录(如黑名单)
out_dir 清洗后的文件输出目录
raw_dir 待处理文件所在mull
dirty_dir 存储清洗出来的脏数据,如为空则不存
:--------------- :-------------------
split_multi_repost 将微博转发数据按"//@aaa XXXX //@bbb XXX"撕开成多句
no_utter_dup 如果 context == response 则去掉该对话
re_name 人名用 , ...替换
no_ad 去除可能是广告的对话(同样的回复对应多个context)借鉴论文
de_generic_dialog 去通用回复 借鉴论文
no_short_response 去掉对话尾部所有过短回复
:--------------- :-------------------
bert_clean 使用BertTokenizer 中函数清理句子
cleantext_clean 使用clean-text 清理 (电话号、邮箱、unicode错误等)
:--------------- :-------------------
no_short 去除过短的句子
no_long 去除过长的句子
de_reply_tag 去除微博中 "回复 @XXX:"
de_hashtag 去除句中 "# XXX#"
de_emotion 去除句中 ": XXX:"
de_mention 去除句子中 "@Cindy", "@Bob:", "@Amy:" 等
no_mention 去除包含 @XXX 的句子
de_repost 去除句中 "//XXX"
de_duplicated 句中短语降重 (待用后缀算法优化)
de_emoji 去除emoji (代补全)
no_special_topic 过滤包含特定名单词的对话对话
no_str_blacklist 过滤包含黑名单词的对话
no_toupiao 判断是否是微博投票
no_specific_utter 删除一些特定句子
contain_zh 删掉不包含中文的句子
de_single_repost_mention 去掉 "@XXX:"
de_weibo_url 去除 http:\t.c
de_url 去除 url
de_angle 去除 其中XX为非中文
de_alpha_num 去除长串无意义的数字字母组合
de_specific 去除句中固定pattern
:--------------- :-------------------
de_showall 去除某些特定文件中的 "...显示全部"
de_brackets 去除某些特定文件中的 "[XXX]"
:--------------- :-------------------
no_word_blacklist 过滤分此后的黑名单词的对话
no_alpha_noise 过滤掉含有不成 英文单词的 字母组合 的句子
check_confuse_word 保存包含混淆名单词的对话进行recall
yda_dedupl 如果一个词语在句子中出现的比例 超过一个阈值则放弃该句子
Owner
Yida
Yida
Toward Model Interpretability in Medical NLP

Toward Model Interpretability in Medical NLP LING380: Topics in Computational Linguistics Final Project James Cross ( 1 Mar 04, 2022

🎐 a python library for doing approximate and phonetic matching of strings.

jellyfish Jellyfish is a python library for doing approximate and phonetic matching of strings. Written by James Turk James Turk 1.8k Dec 21, 2022

Sorce code and datasets for "K-BERT: Enabling Language Representation with Knowledge Graph",

K-BERT Sorce code and datasets for "K-BERT: Enabling Language Representation with Knowledge Graph", which is implemented based on the UER framework. R

Weijie Liu 834 Jan 09, 2023
Edge-Augmented Graph Transformer

Edge-augmented Graph Transformer Introduction This is the official implementation of the Edge-augmented Graph Transformer (EGT) as described in https:

Md Shamim Hussain 21 Dec 14, 2022
Transformers Wav2Vec2 + Parlance's CTCDecodeTransformers Wav2Vec2 + Parlance's CTCDecode

🤗 Transformers Wav2Vec2 + Parlance's CTCDecode Introduction This repo shows how 🤗 Transformers can be used in combination with Parlance's ctcdecode

Patrick von Platen 9 Jul 21, 2022
IMDB film review sentiment classification based on BERT's supervised learning model.

IMDB film review sentiment classification based on BERT's supervised learning model. On the other hand, the model can be extended to other natural language multi-classification tasks.

Paris 1 Apr 17, 2022
Simple, hackable offline speech to text - using the VOSK-API.

Simple, hackable offline speech to text - using the VOSK-API.

Campbell Barton 844 Jan 07, 2023
使用pytorch+transformers复现了SimCSE论文中的有监督训练和无监督训练方法

SimCSE复现 项目描述 SimCSE是一种简单但是很巧妙的NLP对比学习方法,创新性地引入Dropout的方式,对样本添加噪声,从而达到对正样本增强的目的。 该框架的训练目的为:对于batch中的每个样本,拉近其与正样本之间的距离,拉远其与负样本之间的距离,使得模型能够在大规模无监督语料(也可以

58 Dec 20, 2022
CCF BDCI 2020 房产行业聊天问答匹配赛道 A榜47/2985

CCF BDCI 2020 房产行业聊天问答匹配 A榜47/2985 赛题描述详见:https://www.datafountain.cn/competitions/474 文件说明 data: 存放训练数据和测试数据以及预处理代码 model_bert.py: 网络模型结构定义 adv_train

shuo 40 Sep 28, 2022
sangha, pronounced "suhng-guh", is a social networking, booking platform where students and teachers can share their practice.

Flask React Project This is the backend for the Flask React project. Getting started Clone this repository (only this branch) git clone https://github

Courtney Newcomer 17 Sep 29, 2021
💥 Fast State-of-the-Art Tokenizers optimized for Research and Production

Provides an implementation of today's most used tokenizers, with a focus on performance and versatility. Main features: Train new vocabularies and tok

Hugging Face 6.2k Dec 31, 2022
Generate product descriptions, blogs, ads and more using GPT architecture with a single request to TextCortex API a.k.a Hemingwai

TextCortex - HemingwAI Generate product descriptions, blogs, ads and more using GPT architecture with a single request to TextCortex API a.k.a Hemingw

TextCortex AI 27 Nov 28, 2022
DELTA is a deep learning based natural language and speech processing platform.

DELTA - A DEep learning Language Technology plAtform What is DELTA? DELTA is a deep learning based end-to-end natural language and speech processing p

DELTA 1.5k Dec 26, 2022
This repository contains data used in the NAACL 2021 Paper - Proteno: Text Normalization with Limited Data for Fast Deployment in Text to Speech Systems

Proteno This is the data release associated with the corresponding NAACL 2021 Paper - Proteno: Text Normalization with Limited Data for Fast Deploymen

37 Dec 04, 2022
Conditional Transformer Language Model for Controllable Generation

CTRL - A Conditional Transformer Language Model for Controllable Generation Authors: Nitish Shirish Keskar, Bryan McCann, Lav Varshney, Caiming Xiong,

Salesforce 1.7k Dec 28, 2022
Ptorch NLU, a Chinese text classification and sequence annotation toolkit, supports multi class and multi label classification tasks of Chinese long text and short text, and supports sequence annotation tasks such as Chinese named entity recognition, part of speech tagging and word segmentation.

Pytorch-NLU,一个中文文本分类、序列标注工具包,支持中文长文本、短文本的多类、多标签分类任务,支持中文命名实体识别、词性标注、分词等序列标注任务。 Ptorch NLU, a Chinese text classification and sequence annotation toolkit, supports multi class and multi label classifi

186 Dec 24, 2022
The proliferation of disinformation across social media has led the application of deep learning techniques to detect fake news.

Fake News Detection Overview The proliferation of disinformation across social media has led the application of deep learning techniques to detect fak

Kushal Shingote 1 Feb 08, 2022
Geometry-Consistent Neural Shape Representation with Implicit Displacement Fields

Geometry-Consistent Neural Shape Representation with Implicit Displacement Fields [project page][paper][cite] Geometry-Consistent Neural Shape Represe

Yifan Wang 100 Dec 19, 2022
Tutorial to pretrain & fine-tune a 🤗 Flax T5 model on a TPUv3-8 with GCP

Pretrain and Fine-tune a T5 model with Flax on GCP This tutorial details how pretrain and fine-tune a FlaxT5 model from HuggingFace using a TPU VM ava

Gabriele Sarti 41 Nov 18, 2022
Question answering app is used to answer for a user given question from user given text.

Question answering app is used to answer for a user given question from user given text.It is created using HuggingFace's transformer pipeline and streamlit python packages.

Siva Prakash 3 Apr 05, 2022