Research on Traffic Accident Causation Based on Visual Analysis
The study of traffic accident characteristics and causation analysis is the core problem in traffic safety analysis.The existing mainstream methods of Apriori-based traffic accident mining analysis process will generate a large number of invalid rules,which will interfere with the causation analysis and affect the analysis efficiency.In this paper,it combines visualization with associ-ation rule mining methods to visually construct accident mining constraint sets and guide users to set reasonable mining constraints.Using the traffic accident data of California in 2019 as the research object,the visual analysis of traffic accidents is conducted,the main features of accidents are extracted,the accident mining constraint sets are constructed,and the optimized Apriori is used to mine the deep causative factors of accidents.The experiments show that the method can effectively reduce the generation of invalid rules and improve the mining efficiency.
万方数据
维普
