X.H. Wang

Pages: 37-46

Abstract
Accurate traffic accident classification is of great significance to realize the analysis, treatment and prevention of traffic accidents. In order to improve the accuracy of traffic accident data classification, this paper designs a traffic accident data classification algorithm based on decision tree. Firstly, by determining the participation rate of both parties in traffic accidents, the moments of the first three stages are used to characterize the data characteristics after the accident, and the data characteristics of traffic accidents are collected; Then, the positive samples are set by random linear interpolation method, and the relative stability of data features is controlled by logistic regression model; Finally, the maximum information gain is used to segment the characteristic attributes of traffic accident data. The feature classifier of traffic accident data is designed through decision tree, and the error of classification result is corrected to realize classification. The experimental results show that the highest accuracy of this method for traffic accident data classification is about 98%.
Keywords: decision tree; traffic accident data; moment of the first three stages; logistic regression model; regression model