Study Of China's Stock Trading Forecast Based On Social Media Information

With the rapid development of economy,China's stock market has been fully developed and attracts thousands of investors.However,high profits are always accompanied with high risks,and investors are in urgent need of means of effective stock trading forecast to maximize the benefits and reduce risk.At the same time,the traditional communication habits had been greatly changed by the Internet,as one can quickly get the information he needs through the net and post immediately what he sees and feels onto Weibo or We Chat or other social media websites.It is of great practical significance to crawl the social media information and apply it to forecast the stock market transactions,which is the hotspot in the Big Data Era.BP neural network,which has nonlinear mapping ability,generalization ability and fault tolerance ability,is quite popular in academia in recent years.It is the first choice of neural network in stock forecast analysis.The analysis and evaluation of stock investment behavior by BP neural network can help us understand the law and internal mechanism of stock price operation and correctly forecast the trend of stock trading,so as to take appropriate measures to maintain the stability of stock market and promote the healthy development of China's economy.On the basis of reviewing the related literature,this thesis collects data from Weibo and We Chat,the two most popular social medias in China,and crawls some quantitative indicators such as “forward”,“comment”,“like” and Reading Amount,etc.through the data mining technology.In addition,this thesis analyzes the specific social behavior of investors on Weibo and We Chat,and deeply studies the relationship between Weibo indicators,We Chat indicators and stock trading indicators.Finally,the BP neural network is employed to build the stock trading forecasting model and several different stocks are being examined afterwards.It is of great significance to the study of the relationship between social media and stock market.In this thesis,the author finds that the Weibo indicators and the We Chat indicators have strong correlation with the stock trading information both of the day and of the next day.The correlation between the data from the social media and the stock trading volume and transaction amount is especially obvious.Meanwhile,the stock trading behavior of investors also has a certain relationship with the social behavior reflected in social media.So the enterprises could predict the stock market from the perspective of investors and provide them with targeted investment advices.In addition,with the BP neural network stock trading forecasting model of this thesis,the social media data such as Weibo indicators and We Chat indicators could be directly used to forecast the stock trading volume and transaction amount.Although the forecast based on We Chat is less effective and stable than that of Weibo in general,the actual prediction effect by the social media is still much better than the traditional forecast method based on historical data.Therefore,to a certain extent,social media information is able to predict the changes in stock market trading volume and transaction amount.It can be considered that this thesis provides a certain practical significance to the actual operation of the stock market and to the market investors and regulators.