Data Mining And Characteristic Analysis Of Traffic Accidents Of Long Downhill Tunnels On Mountain Expressways

Last few years,China has made great achievements in the field of road traffic safety control,but due to the large vehicle base and uneven road conditions in China,the number of traffic accidents is still in a leading position in the world,especially in mountainous highways with poor geographical conditions Road tunnels have long downhill sections.Traffic accidents are very densely distributed.In the event of a traffic accident,the consequences are more serious and pose a huge threat to the lives and property of the people.Therefore,the traffic safety management and prevention of mountain highways gradually came into the eyes of the public.Therefore,this paper based on the characteristics analysis and cause analysis of the traffic accident data of long downhill tunnels in mountainous areas,and then carried out research on the frequent occurrence points of mountainous highway accidents,and obtained road accident black spots for future researchers to conduct road traffic safety The exploration of security technology provides scientific basis and data support.This article first analyzes and explains the characteristics of road traffic accident data,including data sources and data types,and explains the main problems of existing road traffic accident data,mainly data integrity and objectivity issues;research road traffic accident data mining The feasibility,including the concept,characteristics and functions of data mining,and the objects and ways of road traffic accident mining.Secondly,based on the statistical analysis of the traffic accident data collected in the Baoding section of the Zhangshi Expressway,the time,space,accident form,vehicle type and road line distribution characteristics of the long downhill tunnel section of the mountainous expressway are obtained.Then,based on the decision tree model,the cause of traffic accidents on long downhill tunnels in mountainous areas was studied;then,the basic methods for identifying accident-prone points were briefly described,and the advantages and disadvantages of each method and the scope of use were analyzed.Due to the shortcomings of the basic method,According to the characteristics of road traffic accidents and the status of the existing data mastered in this research,the new improved cumulative frequency curve method is used to obtain the cumulative frequency scatter diagram of the number of equivalent accidents,and the scatter diagram is placed in the Origin software for graphics Fitting to get the fitting equation.The number of equivalent accidents is specified as a critical value to obtain accident-prone points and potential accident-prone points.Taking the traffic accident data accidents of Baoding section of Hebei Zhangshi Expressway as the original data,the original and improved cumulative frequency curve method was used for empirical analysis.The results show that the improved cumulative frequency curve method is more scientific and The conclusions drawn are more representative.