Research On Causes Analysis And Severity Forecast Of Traffic Accidents In National Trunk Highway Of Jilin Province

Traffic road safety condition is closely related to the public travel safety,countries also attaches great importance on traffic,so in recent years our country road mileage and car ownership increased at the same time,the corresponding improvement measures,to effectively control the growth in the number of traffic accidents,but because of the large population base,the number of traffic accidents is still high.In China,the traffic problem of national and provincial trunk lines is becoming more and more prominent,and every traffic accident will cause no small loss.Now,how to reduce the occurrence of traffic accidents has become a research topic that is very worth exploring.In recent years,with the rapid development and construction of highways in China,the role of national and provincial trunk highways has become more and more prominent,becoming the most important in each five-year plan of provinces and cities.As an important passageway connecting rural highways and expressways,and towns and cities,national and provincial trunk highways play a connecting role in the whole transportation system and have a function of distribution to a certain extent.The main purpose of the study on the traffic safety of national and provincial trunk lines is to prevent the frequent occurrence of traffic accidents and reduce the loss caused by traffic accidents.This paper discusses the occurrence of traffic accidents of national and provincial trunk lines in Jilin Province from two aspects: the main influencing factors of traffic accidents and the prediction of the severity of traffic accidents.The research contents are as follows:Firstly,this paper analyzes the status quo of road traffic safety at home and abroad by using literature review and relevant data statistics methods,and studies the influencing factors of road traffic accidents at home and abroad as well as the status quo of road traffic accident severity prediction research.Secondly introduces the Jilin province at present stage of provincial trunk road traffic characteristics to conduct a comprehensive analysis in detail,from the perspective of people,vehicles,road and environment combined with previous research experience and the formation cause of road traffic accident made a comprehensive and objective analysis,and effective information extraction and remove the invalid information in the data,the groundwork for the below.Thirdly,the data of 13,059 road traffic accidents collected in Jilin Province from 2014 to 2019 are preprocessed,the advantages and disadvantages of association rules and related algorithms are introduced,and the appropriate Apriori algorithm is selected to introduce the relevant concepts,characteristics and steps of Apriori,and the original algorithm is improved.Through Python software implementation,analyze and mine the pre-processed accident data,find out the implicit connection between various factors from the data,get meaningful association rules,and put forward relevant countermeasures.Then,according to the results of the above selected for Jilin province main factors influencing the bigger of the two,and the interaction due to various factors to predict the severity of traffic accidents,it is divided into four levels,build LSTM algorithm based prediction system of traffic accident severity,and compared with the classical BP neural network to prove its rationality.Finally,according to the conclusions,the relevant preventive countermeasures are put forward from four aspects: human,vehicle,road and environment.