基于轨迹数据的大规模路网交通拥挤时空关联规则挖掘

Spatio-temporal Association Rule Mining of Traffic Congestion in a Large-scale Road Network Based on Trajectory Data

提出了K近邻RElim(K neighbor-RElim,KNR)算法和时序K近邻RElim(sequential KNbrRElim,SKNR)算法,利用大规模路网的车辆轨迹数据来挖掘路段拥挤关联规则和拥挤传播时空关联规则。其中KNR算法在RElim算法基础上拓展了空间拓扑约束,可高效从大规模车辆轨迹数据集中挖掘路网中关联性拥挤易发路段,并量化这些路段间拥挤的关联性强度。而SKNR算法进一步以滑动窗口的形式拓展时间维度,可以挖掘出大规模路网中难以直接观测的拥挤传播现象,并追溯拥挤传播路径。以成都路网和车辆轨迹数据的挖掘结果对所提出的算法进行了说明和验证,结果表明了算法的有效性和鲁棒性。

A K neighbor-RElim(KNR)algorithm and a sequential KNbr-RElim(SKNR)algorithm are proposed to mine traffic congestion association rules and congestion propagation spatio-temporal association rules by vehicle trajectory data in a large-scale road network.The KNR algorithm extends the spatial topology constraint based on the RElim algorithm.The KNR can be used to mine the road links prone to congestion from the large-scale trajectory dataset in a large-scale road network and quantify the strength of association for congested road links.The SKNR algorithm expands the time dimension in the form of sliding window and can be applied for mining the congestion propagation phenomenon which is difficult to observe directly in a large-scale road network and tracing the path of congestion propagation.The algorithms are illustrated and verified by the empirical results of the Chengdu road network with vehicle trajectory data.The results show the effectiveness and robustness of the proposed algorithms.