Stock trends prediction combining the public opinion analysis

Stock market is complex and volatile, the stock price trends is not only related to the stock price history, but also have a certain correlation with the national policy, the economic environment and the unexpected events of company. Most of the existing stock market prediction methods are only based on the history of stock price and neglect the influence of public sentiment. Nowadays, Social networks such as micro-blog are developing rapidly and will greatly affect the behavior of investors. In this paper we present a system that combined the existing stock prediction method based on technical indicator with the public opinion analysis, and introduce the definition of micro-blog influence factors. The experiments on The Shanghai Composite Index show that this method can achieve higher accuracy than a single source system.