Data and code underlying the online perceived risk study

This study explores the dynamic nature of perceived risk in vehicle interactions. Using a novel simulation-based approach, time-continuous perceived risk data were obtained from over 140k ratings by 2,164 participants across various traffic scenarios. This package contains the collected perceived risk data, corresponding kinematic data of the simulation, code and software implementation in MATLAB, Python, SPSS, and Origin for data processing, statistical analysis, model calibration, perceived risk prediction, SHAP analysis and visualisation.本研究探究了车辆交互场景下感知风险的动态特性。本研究采用一种新颖的基于仿真的研究方法，从2164名参与者在多种交通场景下的14万余条评分中，获取了时间连续的感知风险数据集。本数据包涵盖所采集的感知风险数据、仿真对应的运动学数据，以及用于数据处理、统计分析、模型校准、感知风险预测、SHAP分析与可视化的MATLAB、Python、SPSS及Origin代码与软件实现。