Prediction of stock market trends has been an area of great interest both to researchers attempting to uncover the information hidden in the stock market data and for those who wish to profit by trading stocks. The extremely nonlinear nature of the stock market data makes it very difficult to design a system that can predict the future direction of the stock market with sufficient accuracy. This work presents a data mining based stock market trend prediction system, which produces highly accurate stock market forecasts. The proposed system is a genetic algorithm optimized decision tree-support vector machine (SVM)
hybrid, which can predict one-day-ahead trends in stock markets. The uniqueness of the proposed system lies in the use of the hybrid system which can adapt itself to the changing market conditions and in the fact that while most of the attempts at stock market trend prediction have approached it as a regression problem, present study converts the trend prediction task into a classification problem, thus improving the prediction accuracy significantly. Performance of the proposed hybrid system is validated on the historical time series data from the Bombay stock exchange sensitive index (BSE-Sensex). The system performance is then compared to that of an artificial neural network (ANN) based system and a naïve Bayes based system. It is found that the trend prediction accuracy is highest for the hybrid system and the genetic algorithm optimized decision treeSVM hybrid system outperforms both the artificial neural network and the naïve bayes based trend prediction systems.
Dr. Binoy B. Nair, Mohandas, V. P., and Dr. Sakthivel N.R., “A genetic algorithm optimized decision tree-SVM based stock market trend prediction system”, International Journal on Computer Science and Engineering, vol. 2, pp. 2981–2988, 2010.