Forecasting the Trends of Stock Price through Social Networks by Fuzzy Support Vector Machine with Possibility Measures

Abstract

Obviously, there are many advantages for us to correctly forecast the price trends of stock. According to the hypothesis of efficiency market, the prices of stock are evaluated by all the current useful information. The mood of society plays an important role that affects the trend of current stock market price. The entire social mood corresponding to a given company is a critical factor which influences the stock price for that company. In recent years, the popularity of online social networks provides huge number of available social mood data. Hence, mining information from social networks and historical stock prices can improve the ability of predicting trend in stock market.In this study, we extracted the hidden topic model and emotional information from user's posts in social networks. Besides, we developed a fuzzy support vector machine to merge the abundant information from the on-line posts, which can be used to forecast the trend of stock prices. Fuzzy set theory is very useful for this study because the texts are fuzzy in itself (such as high/low and big/small), and there is an ambiguous boundary between rise and fall categories. For example, going up either 10% or 1% belongs to rise category, but is different in degree. In comparison with traditional support vector machine, the method proposed in this study is significantly better than the forecasting model of traditional support vector machine.