Abstract:Traffic congestion is becoming an increasingly serious problem. Traffic accidents caused by risky driving behaviors are one important cause. Therefore, the accurate evaluation of driving behaviors has become a research hotspot. This study puts forward an evaluation algorithm of driving behaviors based on the combination of FCM and BP neural network. Firstly, FCM is used to make initial clusters of driving behaviors. Secondly, in accordance with the results of clusters, an algorithm that the typical samples are automatically selected as the training samples for BP neural network classifier is proposed. Finally, the trained BP neural network is used to classify the driving behaviors. The research result shows that the algorithm can eliminate subjective factors and make accurate, objective and efficient driving behavior evaluation.