Comparison of road traffic accident prediction effects based on SVR and BP neural network

Road traffic accident prediction is an important topic in traffic research. Taking China's traffic safety status as the research object, according to the non-linear characteristics of China's road traffic accidents, using principal component analysis to reduce the dimensionality of road traffic accident statistics data in China, comparing the two methods of SVR and BP neural network Comparison shows that the estimation accuracy of SVR is better than that of BP neural network, and its prediction accuracy meets the requirements of road traffic accident prediction.