【摘要】 随着互联网的发展,人们更愿意在网络上分享自己对热点事件的观点并发表自己的评论,这些评论通常包含了个人的情绪和情感倾向,所以对网络短评进行情感分析能高效、精准的挖掘人们的情感态度。本系统首先对知乎平台的网络短评进行定向抓取,然后进行数据清洗、分析出人们的情感取向,最后利用词云进行直观的展示。本系统的情感分析模块主要使用了Google的TensorFlow框架,并采用长短期记忆网络(LSTM)对已经标注好正负情感评论的语料进行训练,然后使用PythonScrapy框架对知乎热榜评论进行定向数据爬取,并做出情感预测,最后使用Tornado框架实现情感分析系统的Web图形化操作。利用本系统可以高效、精准的挖掘人们的情感态度,有助于舆情分析、用户分析等方面的应用。更多还原

【Abstract】 With the development of the Internet, people are more willing to share their views on hot issues and make their own comments on the Internet, which usually contain personal emotions and emotional tendencies. Therefore, the emotional analysis of online short comments can effectively and accurately explore people’s emotional attitudes. This system first grabs the network short comments on zhihu platform, then cleans the data, analyzes people’s emotional orientation, and finally USES the word cloud to display intuitively. The emotional analysis module of this system mainly USES the Google TensorFlow framework, using both short-term and long-term memory network(LSTM) to have training corpus with the positive and negative emotional comments, and then use the Python Scrapy framework to zhihu crawl hot list of comments for directional data, and make emotional prediction, sentiment analysis is realized by using Tornado framework Web graphical operation of the system. This system can be used to effectively and accurately explore people’s emotional attitude, which is conducive to the application of public opinion analysis, user analysis and other aspects.更多还原