Skip to content

Social media (Weibo) comments analyzing toolbox in Chinese 微博评论分析工具, 实现功能: 1.微博评论数据爬取; 2.分词与关键词提取; 3.词云与词频统计; 4.情感分析; 5.主题聚类

License

Notifications You must be signed in to change notification settings

KimMeen/Weibo-Analyst

Repository files navigation

About This Repository

Please follow the restrictions in Apache-2.0 Licence before you refer to this repository.

@Ming Jin (mingj2@student.unimelb.edu.au)

This repository is a simple NLP project for beginners and will be updated occasionally.

Environment: Python3.6

Pull requests are always welcome!

To-do list:

  1. Sentiment analysis can be done by using LSTM.
  2. Crawler can be improved based on some libraries like BeautifulSoup.

Introduction

这是一个微博评论分析工具,实现功能主要有:

  1. 微博评论数据爬取
  2. 分词与关键词提取
  3. 词云与词频统计
  4. 情感分析
  5. 主题聚类

正常状态下实现效果在: “ 案例:泰国大象踩踏伤人事件 ” 注意:案例中最后表格需要自己根据LDA结果进行统计

This is a Weibo comments processing toolbox, which has been implemented for:

  1. Weibo comments crawler that based on regular expression
  2. Tokenization, filtration and key words extraction
  3. Words cloud and visualization
  4. Sentiment analysis
  5. Topic clustering that based on LDA

Pre-Requirements Checklist

MySQL is required (e.g., You may use MySQL Workbench)

  1. importlib
  2. sys
  3. time
  4. requests
  5. lxml
  6. pymysql
  7. jieba
  8. PIL
  9. wordcloud
  10. snownlp
  11. logging
  12. configparser
  13. random
  14. codecs

Citing The Code

BibTex reference format:

@misc{WeiboAnalyst,
title={Weibo-Analyst: An Open-Source Python Library for Social Media Comments Analysis},
url={https://github.com/KimMeen/Weibo-Analyst},
author={Ming Jin},
year={2018}
}

About

Social media (Weibo) comments analyzing toolbox in Chinese 微博评论分析工具, 实现功能: 1.微博评论数据爬取; 2.分词与关键词提取; 3.词云与词频统计; 4.情感分析; 5.主题聚类

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages