Traffic Accident Risk Factor Identification Based on Complex Network

Abstract Identifying the risk characteristics of traffic accidents is of great significance for in-depth understanding and promotion of traffic safety. 85 serious traffic accidents were selected from the road traffic accident liability certification and serious accident reports issued by the traffic police department, and 110 risk points and 218 triggering relationships were abstracted based on network theory analysis to establish a traffic accident risk network. Analyse and compare indicators such as network density and network diameter to reveal the overall nature of the traffic accident risk network. The key risk factors of the risk network are analysed and identified using node indicators such as degree and betweenness. The results show that most traffic accidents are triggered by human factors, among which the driver factor is the main factor. Speeding and improper emergency measures are the two most important factors that need to be controlled. Based on the results of this analysis, this article puts forward management suggestions to improve the level of traffic safety.