{"title":"Analysis of bus-stop operating state based on GPS data","authors":"Hongtao HUANG, Mei XIAO, Qian LIU, Xiuling MING, Haoyi BIAN","doi":"10.3724/sp.j.1249.2023.03326","DOIUrl":null,"url":null,"abstract":"中国西安市公交车全球定位系统轨迹数据为例,建立平均服务时间和服务车数特征参数反映公交站的运行 状态,并通过分析站点内公交车辆速度、里程及加速度之间关系计算站台服务时间.使用Hopkins统计量 和轮廓系数分析可聚性和聚类数,结合高斯混合模型(Gaussian mixture model, GMM)对公交站运行状态进 行识别分类.构建 SMOTEENN-XGBoost(synthetic minority oversampling technique edited nearest neighbours extreme gradient boosting)站点运行状态预测模型 , 引入可解释机器学习框架 SHAP(Shapley additive explanation)分析站台属性、道路及环境对模型的影响.结果表明,公交站运行状态可分为3类,类型I的 平均服务时间最长,类型II的平均服务时间和服务车数最少,类型III的服务车数最多;所建立 SMOTEENN-XGBoost 模型的准确率为 94. 68%,精确率为 94. 69%,召回率为 91. 04%,F1分数为 92. 26%, 与极限梯度提升(extreme gradient boosting, XGBoost)、 逻辑回归(logistic regression, LR)、 随机森林 (random forest, RF)、梯度提升决策树(gradient boosting decision tree, GBDT)和 k 近邻(k-nearest neighbors, KNN)5种模型对比,本模型能够精准预测站点运行状态;对站点运行状态具有影响作用的因素按照重要程","PeriodicalId":35396,"journal":{"name":"Shenzhen Daxue Xuebao (Ligong Ban)/Journal of Shenzhen University Science and Engineering","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Shenzhen Daxue Xuebao (Ligong Ban)/Journal of Shenzhen University Science and Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3724/sp.j.1249.2023.03326","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Engineering","Score":null,"Total":0}
引用次数: 0