Smart Grid and Internet of Things. Third EAI International Conference, SGIoT 2019, TaiChung, Taiwan, December 5-6, 2019, Proceedings

Research Article

Prediction Traffic Flow with Combination Arima and PageRank

Download
150 downloads
  • @INPROCEEDINGS{10.1007/978-3-030-49610-4_11,
        author={Cheng-fan Li and Jia-xin Huang and Shao-chun Wu},
        title={Prediction Traffic Flow with Combination Arima and PageRank},
        proceedings={Smart Grid and Internet of Things. Third EAI International Conference, SGIoT 2019, TaiChung, Taiwan, December 5-6, 2019, Proceedings},
        proceedings_a={SGIOT},
        year={2020},
        month={6},
        keywords={Network structure Traffic congestion Armia model},
        doi={10.1007/978-3-030-49610-4_11}
    }
    
  • Cheng-fan Li
    Jia-xin Huang
    Shao-chun Wu
    Year: 2020
    Prediction Traffic Flow with Combination Arima and PageRank
    SGIOT
    Springer
    DOI: 10.1007/978-3-030-49610-4_11
Cheng-fan Li,*, Jia-xin Huang1,*, Shao-chun Wu1,*
  • 1: Shanghai University
*Contact email: lchf@shu.edu.cn, 1204833945@qq.com, scwu@shu.edu.cn

Abstract

Modern traffic network information is similar to the complex network structure in that the links between the sections are quite complex. Therefore, predicting the traffic flow between sections can effectively relieve traffic congestion. To solve this problem, this paper proposes a combined model of Arima and PageRank to predict the traffic flow of each section of the road network. First, the trained Armia model is used to predict the average speed and traffic flow of each section, and then the PageRank model is used to calculate the weight of each section. The product of traffic flow and weight is output as the final result. Through the experiment of highway traffic data in PeMS database, this method is verified to be able to predict the traffic flow of the whole road network.