基于车联网数据挖掘的驾驶员行为分析

Driver Behavior Analysis Based on Vehicle Network Data Mining

随着汽车数量持续增加,车与路的矛盾日益突出,由此引发了交通运输和管理问题.而智能交通技术和车联网技术是解决交通拥堵和管理问题的有效途径.因此,运用数据挖掘的方法,使用isolation forest算法和SOM算法对车联网技术采集的交通数据进行预处理并提取特征值,再通过K-mesns聚类算法归类并添加标签,最后通过BP神经网络构建分类器,实现了对驾驶员驾驶行为的分类和评价,实例计算结果表明,该算法能客观、有效的评价驾驶员驾驶行为,对交管部门、汽车营运企业有具有一定参考价值.

As the number of cars continues to increase,the contradiction between cars and roads has become increasingly prominent,which has led to transportation and management problems.Intelligent transportation technology and vehicle networking technology are effective ways to solve traffic congestion and management problems.Therefore,using the data mining method,using the algorithm and algorithm to preprocess the traffic data collected by the Internet of Vehicles technology and extract the feature values,then classify and add tags through the clustering algorithm,and finally construct the classifier through the neural network,and realize the classification and evaluation of driver's driving behavior,the example calculation results show that the algorithm can objectively and effectively evaluate the driver's driving behavior,and has certain reference value to the traffic control department and automobile operation enterprise.