In this study, a hybrid data mining methodology, rough set based support vector machine (RS-SVM) model, is proposed to explore stock market tendency. In this approach, rough set is used for feature vectors selection to reduce the computation complexity of SVM and then the SVM is used to identify stock market movement tendency based on the historical data. To evaluate the forecasting ability of RS-SVM, we compare its performance with that of conventional methods and neural network models. The empirical results reveal that RS-SVM outperforms other forecasting models, implying that the proposed approach is a promising model to stock market tendency exploration.