The invention discloses a traffic accident prediction system based on Ada-XGBoost. The system comprises a base prediction model module and an actual prediction application module. The method mainly comprises the following steps: S1, collecting traffic accident historical data, and arranging the traffic accident historical data into a training sample set; S2, performing parameter optimization on the XGBoost by applying a grid search method in combination with K-fold cross validation; S3, integrating the optimized XGBoost on the basis of the AdaBoost; S4, performing parameter optimization on AdaBoost according to the method in S2; S5, constructing an AdaBoost-XGBoost strong predictor (Ada-XGBoost), training the model, and calculating performance evaluation indexes; S6, judging the rationality of the evaluation indexes, and if the evaluation indexes are reasonable, storing the model as a base prediction model module; otherwise, repeating the steps S2 to S5 until rationality is achieved; S7, inputting the to-be-tested sample set into the actual prediction application module to realize traffic accident prediction. According to the invention, the XGBoost with good prediction performance is taken as a base algorithm, an AdaBoost integrated framework is introduced, parameter optimization is carried out, and the constructed traffic accident prediction system is easy to implement, high in efficiency and strong in generalization ability.

    本发明公开了一种基于Ada‑XGBoost的交通事故预测系统,该系统包括基预测模型模块和实际预测应用两个模块。主要包括以下步骤:S1、收集交通事故历史数据,整理成训练样本集;S2、应用网格搜索法结合K折交叉验证对XGBoost进行参数寻优;S3、基于AdaBoost集成优化后的XGBoost;S4、同S2的方法,对AdaBoost进行参数寻优;S5、构建AdaBoost‑XGBoost强预测器(Ada‑XGBoost),训练模型并计算性能评价指标;S6、判断评价指标的合理性,如果合理,保存模型为基预测模型模块。否则,重复S2至S5,直至合理;S7、将待测样本集输入实际预测应用模块,实现交通事故预测。本发明以预测性能较佳的XGBoost为基算法,引入AdaBoost集成框架,同时进行参数寻优,构建的交通事故预测系统易于实现、效率较高、泛化能力较强。


    Access

    Download


    Export, share and cite



    Title :

    Traffic accident prediction system based on Ada-XGBoost


    Additional title:

    一种基于Ada-XGBoost的交通事故预测系统


    Contributors:
    CHANG RUNQI (author)

    Publication date :

    2021-08-06


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    The Use of XGBoost Algorithm to Analyse the Severity of Traffic Accident Victims

    I Made Sukarsa / Ni Kadek Dwi Rusjayanthi / Made Srinitha Millinia Utami et al. | DOAJ | 2023

    Free access

    Prediction system for traffic accident

    BACK JU YONG | European Patent Office | 2019

    Free access

    Prediction system for traffic accident

    BACK JU YONG | European Patent Office | 2019

    Free access

    Traffic Data-Empowered XGBoost-LSTM Framework for Infectious Disease Prediction

    Guo, Kehua / Shen, Changchun / Zhou, Xiaokang et al. | IEEE | 2024