The invention provides a big data-driven highway traffic incident duration prediction method. Comprising the following steps: step 1, establishing a highway traffic event data set, and extracting fields required by research; step 2, preprocessing the extracted data fields; 3, on the basis of data preprocessing, establishing a random forest model, and predicting the duration of the highway traffic incident; 4, on the basis of data preprocessing, in combination with the feature vector of the event text information, establishing a text data feature prediction model based on a random forest; and 5, comparing and analyzing the precision of the prediction model by adopting an average absolute error and an average relative error. According to the method, the structured influence factors and the semi-structured text information are fused, and the duration of the highway traffic event can be predicted under different conditions, so that a road manager and a user can more comprehensively and reliably pre-judge the traffic event.
本发明提供了一种大数据驱动的高速公路交通事件持续时间预测方法;包括:步骤1,建立高速公路交通事件数据集,提取研究所需的字段;步骤2,对提取的数据字段进行预处理;步骤3,在数据预处理的基础上,建立随机森林模型,对高速公路交通事件持续时间进行预测;步骤4,在数据预处理的基础上,结合事件文本信息的特征向量,建立基于随机森林的文本数据特征预测模型;步骤5,采用平均绝对误差、平均相对误差对预测模型精度进行对比分析。本发明融合结构化的影响因素和半结构化的文本信息,能够在不同状况下对高速公路交通事件持续时间进行预测,从而实现道路管理者和使用者对交通事件更加全面可靠的预判。
Big data-driven highway traffic incident duration prediction method
一种大数据驱动的高速公路交通事件持续时间预测方法
2022-06-14
Patent
Electronic Resource
Chinese
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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Traffic Incident Duration Prediction Based on K-Nearest Neighbor
British Library Conference Proceedings | 2013
|Competing risks mixture model for traffic incident duration prediction
Online Contents | 2015
|