Feature Recognition of Urban Road Traffic Accidents Based on GA-XGBoost in the Context of Big Data

Yi Qu,Zhengkui Lin,Xiaonan Zhang,Honglei Li

doi:10.1109/access.2019.2952655

Abstract

The identification of the characteristics of urban road traffic accidents is of great significance for reducing traffic accidents and the corresponding losses. In the context of big data, to accurately understand the characteristics of traffic accidents, the feature set of urban road traffic accidents is proposed, the XGBoost model is used to classify traffic accidents into minor accidents, general accidents, major accidents and serious accidents, and a GA-XGBoost feature recognition model is built. The GA-XGBoost feature recognition model is based on the genetic algorithm (GA) as a factor search algorithm and is verified by applying the big data of traffic accidents in a Chinese city from 2006 to 2016; in addition, the model is compared with the GA-RF, GA-GBDT and GA-LightGBM models. The results show that the GA-XGBoost model can accurately identify the features of the traffic accidents in 7 cities, including driving experience, illegal driving behavior, vehicle age, road intersection type, weather conditions, traffic flow and time interval. Compared with the GA-RF, GA-GBDT and GA-LightGBM models, the recognition features are more accurate, and the performance is better.

Highlights

Modern information technology has brought great changes and challenges to road traffic behavior and its safety. ‘‘Internet + traffic’’ and intelligence have become the inevitable development trends for future road traffic systems
In this paper, a feature recognition model based on the big data of traffic accidents is proposed
The model uses the XGBoost algorithm as a multi-classifier of traffic accident levels to identify the characteristics of urban road traffic accidents by a genetic algorithm (GA) search for the optimal solution

Summary

INTRODUCTION

Modern information technology has brought great changes and challenges to road traffic behavior and its safety. ‘‘Internet + traffic’’ and intelligence have become the inevitable development trends for future road traffic systems. By effectively identifying the characteristics of urban traffic accidents and analyzing the causes of traffic accidents can the function of traffic safety service systems be more effectively supported and the construction of intelligent transportation systems be improved. Through big data mining methods, this paper identifies the characteristics of urban road traffic accidents with important economic and social significance, which is conducive to reducing and avoiding traffic accidents and fundamentally improving road traffic safety and provides theoretical support for the construction of intelligent traffic systems. Often use traditional statistical methods, such as questionnaire surveys and regression analysis, to identify the characteristics of road traffic accidents, so that a large amount of traffic accident data cannot be utilized to improve traffic safety. The GA-XGBoost model, which uses XG-Boost to classify the traffic accident severity and the genetic algorithm to search traffic accident characteristics, is constructed to identify the characteristics of urban road traffic accidents based on traffic accident big data and provides effective theories and tools for urban road traffic management

DEFINITION OF THE GRADE AND FEATURES OF ROAD TRAFFIC ACCIDENTS

EMPIRICAL ANALYSIS

DATA SOURCES

Findings

CONCLUSION