基于微博的股票投资者未来情感倾向识别研究

Method to Identify the Future Stock Investor Sentiment Orientation on Chinese Micro-blog

近年来,微博越来越受到网络用户的青睐,成千上万的用户通过发布微博共享他们的观点和情感。其中,有大量带有情感倾向(认为某事物"好"或"坏")的微博,这些微博反映了作者的情绪。投资者情绪(investor sentiment)是研究经济市场走向的重要指标,行为金融学认为股票投资者情绪影响投资者决策,进而影响股票市场,而反映股票投资者情绪的重要指标是投资者对股票市场未来行情的情感倾向(认为股票市场未来行情"好"或"坏")。通过对新浪微博(目前最大的中文微博平台)上股票投资者发布的文本进行情感信息方面的分析与研究,提出了一种自动识别股票投资者未来情感倾向的方法。该方法分为两级识别,第一级是:识别出微博中包含未来情感的句子;第二级是:将第一级识别出来的包含未来情感的句子分为正面评论(看涨)和负面评论(看跌)。实验结果表明,所提方法对自动识别股票投资者的未来情感倾向达到了非常好的效果。

Recently,Micro-blog has attracted more and more interests of internet users.Thousands of the users share their views and opinions through micro-blog.There are a large number of texts with sentiment orientation（thinking something is ＂good＂ or ＂bad＂） on the Micro-blog.These texts reflect the authors’ emotion.Investor sentiment is the important indicator to research on the economic market trends.On behavioral finance,Stock investor sentiment affects the investors’ decision.Then it will affect stock market.The stock investor sentiment orientation（thinking the future market will be ＂good＂ or ＂bad＂） on the future stock market is the indicator to reflect the investor emotion.In this paper,we proposed a method of sentiment classification and apply it to perform sentiment classification on Sina micro-blog（currently,the largest Chinese micro-blog platform）.In detail,our approach contains two-step classifier.Firstly,the first classifier will identify the sentence that contains the future sentiment.Secondly,use the second classifier to classify the sentence which identified by the first classifier into positive or negative.The experimental results show that our method achieves a decent performance on identifying the future stock investor sentiment orientation