现代电子技术2025，Vol.48Issue(14)：50-56,7.DOI:10.16652/j.issn.1004-373x.2025.14.009
基于改进Apriori算法的不良驾驶行为关联分析
Adverse driving behavior correlation analysis based on improved Apriori algorithm
摘要
Abstract
The increasing complexity of bad driving behavior poses a serious threat to road traffic safety.In order to explore the potential law of adverse driving behavior,the driving data of Harbin private cars in the morning and evening rush hours are collected by means of OBD.The python data processing platform is used to identify 4 bad driving behaviors,including speeding,sudden shifting,sharp turning and fast lane changing.An improved Apriori association rule mining algorithm is proposed based on behavioral datasets.The particle swarm optimization(PSO)algorithm is introduced to optimize the two important parameters of support and confidence in Apriori algorithm,and the hash mapping table is used to improve the running efficiency of Apriori algorithm.The experimental results show that the running time of the improved Apriori algorithm is 8.26％and 9.27％higher than that of Apriori on two data sets,respectively.The correlation results show that adverse driving behavior does not exist alone,with the strongest correlation between sharp turns,rapid lane changes and rapid acceleration,followed by overspeed behavior and sudden gear changes.This study can provide reference for the study of driving style and effectively carry out traffic accident prevention and early warning.关键词
驾驶安全/不良驾驶行为/数据挖掘/关联分析/改进Apriori算法/粒子群优化算法Key words
driving safety/adverse driving behavior/data mining/association analysis/improved Apriori algorithm/particle swarm optimization分类
信息技术与安全科学引用本文复制引用
韩锐,于长海,丁庆国,石朋炜..基于改进Apriori算法的不良驾驶行为关联分析[J].现代电子技术,2025,48(14):50-56,7.基金项目
黑龙江省重点研发计划项目:寒区道路交通系统低碳智能化与安全提升技术(JD22A014) （JD22A014）
