X.J. Zhang

Pages: 67-76

Abstract
Road traffic safety is an important part of road traffic system. Reliable and accurate traffic safety prediction model is the basis and difficulty of road traffic safety. Aiming at the problems of poor prediction effect, low accuracy and efficiency in the current urban road traffic safety risk prediction methods, this paper proposes an urban road traffic safety risk prediction method based on RS-SVM combination model. Firstly, according to the attribute reduction theory of rough set and the classification projection method of support vector machine, the RS-SVM combination model is constructed. Then, by calculating the urban road traffic safety risk value, the urban road traffic safety risk prediction index system is constructed. Finally, the RS algorithm is used to screen the risk prediction indicators of urban road traffic safety, and the RS-SVM combination model is used to predict the risk of urban road traffic safety, and the risk prediction results are obtained. The experimental results show that this method has good risk prediction effect and can effectively improve the accuracy and efficiency of risk prediction.
Keywords: RS-SVM combination model; rough set attribute reduction; support vector machine; urban road; traffic safety; risk projection