市政技术2025，Vol.43Issue(5)：1-14,49,15.DOI:10.19922/j.1009-7767.2025.05.001
基于随机森林算法的青年驾驶人风险感知差异性分类及其驾驶行为研究
Research on Classification of Risk Perception Differences and Driving Behavior of Young Drivers Based on Random Forest Algorithm
摘要
Abstract
To address road safety issues of young drivers,a difference-analysis of their perceived risk abilities were conducted from physiological,behavioral,and psychological aspects.A perceived risk classification model was es-tablished by the Random Forest algorithm(RF).Driving simulation experiments were applied to collect characteris-tic indicator data of electrocardiogram(ECG)and electroencephalogram(EEG),as well as driving operations.The optimal feature variables for model input was determined by correlation analysis.Experimental results showed that the proposed model achieved an overall accuracy of 92.5％,precision of 93.0％,recall of 93.0％,and an F1-score of 0.92,superior performance than XGBoost and LightGBM algorithms.Further analysis of risk perception differ-ences across scenarios revealed that young male drivers with high risk perception exhibited a 14.74 ms2 increase of low-frequency power(LF value)before entering implicit scenarios than that of explicit scenarios,indicating insuf-ficient cautiousness;While young female drivers with low-risk perception showed significantly higher physiological indicators(e.g.,heart rate growth rate and LF values)than other groups,reflecting lower cautiousness and poorer driving safety;Additionally,young drivers with moderate-risk perception displayed notable gender differences,namely the averaging driving speeds of young male drivers is 7.06 km/h faster than females,and an 8.07 ms2 reduc-tion in LF values,suggesting males maintain higher cautiousness even during acceleration.These findings provide critical insights for making targeted driver safety education programs and optimizing intelligent assistance systems,ultimately enhancing road safety for young drivers.关键词
交通安全/青年驾驶人/驾驶行为/驾驶人分类/驾驶模拟试验/随机森林算法Key words
traffic safety/young drivers/driving behavior/drivers' classification/driving simulation tests/random forest algorithm(RF)分类
交通工程引用本文复制引用
张遥,焦朋朋,张瑶..基于随机森林算法的青年驾驶人风险感知差异性分类及其驾驶行为研究[J].市政技术,2025,43(5):1-14,49,15.基金项目
国家自然科学基金(52172301) （52172301）
北京市社会科学基金重点项目(21GLA010) （21GLA010）
北京市西城区优秀人才培养资助项目(202338) （202338）
北京建筑大学研究生创新项目(PG2024055) （PG2024055）
