Analysis And Research On Driving Violation Behavior Based On Apriori Algorithm

In today’s society,the number of motor vehicles in China and the number of drivers is increasing,which shows that China has become one of the world’s major motor countries.However,the large number of motor vehicles may lead to a series of traffic safety problems,whose causes are complex and diverse,including road conditions,driving environment,vehicle conditions and driving violations and other factors.Traffic safety concerns people’s livelihood and is closely related to people’s safe travel.According to the existing data,it is found that driving violations are the main factors leading to traffic safety problems.Therefore,it is of great significance to analyze and study driving violations reasonably,dig out the correlation between violations deeply,and provide help to solve traffic safety problems.The specific research contents of this thesis are as follows:According to the characteristics of driving violations,driving speed,standard deviation of speed,vehicle acceleration,standard deviation of acceleration and engine speed were selected as characteristic parameters.Based on principal component analysis,the selected characteristic parameters were transformed into main factors C1 and C2,where C1 had a greater correlation with acceleration and rotational speed,and C2 had a greater correlation with speed.After the driver’s comprehensive score is calculated according to the principal factors,the FCM algorithm is used to conduct cluster analysis on the driver data set and make corresponding evaluation.To solve the problem that the support and confidence of Apriori algorithm need to be set artificially,this thesis proposes an Apriori algorithm based on CA-PSO.The algorithm first seeks for the optimal solution set of fitness function through CA-PSO algorithm,and takes the optimal solution set as the parameter of support and confidence to improve the data mining effect on the data set of the vehicle-carried Internet of Things.The Apriori algorithm based on CA-PSO is used to conduct association rule analysis experiments to mine the potential connection between driving violations and obtain the corresponding association rule results.The experimental results are analyzed and studied to provide management opinions for the traffic management department and put forward safe driving suggestions for drivers.Finally,this paper compares the Apriori algorithm based on CA-PSO,the traditional Apriori algorithm and other improved algorithms respectively from the execution time of the algorithm and the number of association rules obtained by the algorithm.According to the comparative experimental results of this paper,it is confirmed that the improved algorithm in this paper performs faster and the obtained association rules are more effective in the mining of association rules of the vehicle-borne Internet of Things data set.