In recent years, in order to enhance the safety of traffic, intelligent vehicles flourish. To make intelligent vehicles work effectively, driver intention recognition methods have been proposed recently. This study can effectively identify the driver of lane change intention, it explore the different information sources and methods. First, design the different experimental conditions assumed, and through the driving simulator to collect Lane Change (LC) and Lane Keeping (LK) data. Then collect source of information after data processing, created as a part of the training database and the validation database. Then use this database for training and validation of Support Vector Machine (SVM) and Dynamic Bayesian Network (DBN) model. In addition, the relationship between the information sources and the driving intention is analyzed. The experimental results show that SVM with steering wheel angle, yaw angle, yaw rate, and lateral displacement as a source of information, we can effectively identify the driver lane change intention.