Research Of Stock Investment Based On Internet Public Opinion

With the flourish of computer and Internet,investors can obtain almost any information about the listed companies they want through the network.The popularity of we-chat,micro-blog and other social tools makes the Internet evolve from the information publishing platform into a social network containing information dissemination,exchange and interaction.Behavioral finance theory thinks that people are bounded rational,investors make irrational investment decisions due to their sentiment.A large number of studies show that China's stock market is a weak effective market or semi-strong market;Also with the development of more than 20 years,Chinese securities industry will produce massive amounts of information every day,it is difficult for investors to capture the useful data from these massive information;Also,it is difficult to distinguish the authenticity among complex transaction data and related information,leading investors' inefficiency.All of these will lead to investors making unreasonable investment decisions,thus affecting the Chinese stock market.Traditional stock investment analysis using mathematical statistics,with the continuous expansion of data,has been unable to meet the demand.In order to cope with the challenges of large data,emerging stock analysis introduces machine learning,data mining methods,and establishing effective investment models to obtain returns.In this paper,the investors' sentiment influence on investment utility will be derived based on the theory of behavioral finance.Also,this paper will analyze different states of stock market and different investment strategies under various of investor sentiment.Accordingly,the influence of different investment strategies have on the stock market will be elucidated as well.After this,the paper will introduce how investor sentiment(namely Internet public opinion)influences the stock returns.In empirical analysis,the paper will select the network BBS posts as Internet public opinion information source,download stock comments from stocks of Oriental wealth network with the web crawler,and will use Chinese text emotional polarity analysis method to crawl to every post analysis of the emotion,thus building the investor sentiment index and investors' sentiment divergence index based on the Internet public opinion.In this way,this paper will analyze the influence quantitatively.Finally,combining with the neural network algorithm,this paper will forecast the future trends of the stock chosen and make the best investment strategies.The results show that:(1)the data mining algorithm can help to find stocks with excess return rate;(2)investor sentiment will significantly affect the stock return rate;(3)investor sentiment,investor disagreement,trading volume indicators have significant predictive effects on stock returns;(4)when the stock market is on the rise,in the choice of radical investment strategy can get more revenue.