Stock trend forecasting method based on sentiment analysis and system similarity model

This paper combine sentiment analysis based on system similarity model and Bayesian classification model to design a prediction system for the stock plate price trend analysis according to the Internet stock news and information. This system can automatically classified the stock news on the web and apply sentiment analysis to judge related comments and predict the price movements. By the way of cross-rotation test show that the system can effectively predict and analyze the stock market and have good stability.