Forecasting the change of intraday stock price by using text mining news of stock

This paper presents a method for forecasting the change of intraday stock price by utilizing text mining news of stock. This method is based on text mining techniques coupled with rough sets theories and support vector machine classifier. The method can handle without difficulty unstructured news of Taiwan stock market through preprocessing, feature selection and mark. The method also extracts the core phrases by using rough sets theories after the unstructured information has been transformed into structured data. Then, a prediction model is established based on support vector machine classifier. The empirical results show that the proposed model can predict accurately the ups and downs of a stock price within one hour after the news released. The method presented in the study is straightforward, simple and valuable for the short-term investors.