The stock market changes rapidly. Thus, the stock price reversal points play a vital role in investment decisions. Technical analysis can reveal some features of stock price reversal, but to use a sole technical indicator to predict the reversal points can end with sesults that have very low recall and precision rates. In order to improve recall and precision rates, a novel method of using Support Vector Machines (SVM) is proposed to data mine a combination of technical indicators. The experiment results show that this method has a higher recall and precision rate than the original ones.