Collaborative Computing: Networking, Applications and Worksharing. 15th EAI International Conference, CollaborateCom 2019, London, UK, August 19-22, 2019, Proceedings

Research Article

A Next Location Predicting Approach Based on a Recurrent Neural Network and Self-attention

Download
16 downloads
  • @INPROCEEDINGS{10.1007/978-3-030-30146-0_21,
        author={Jun Zeng and Xin He and Haoran Tang and Junhao Wen},
        title={A Next Location Predicting Approach Based on a Recurrent Neural Network and Self-attention},
        proceedings={Collaborative Computing: Networking, Applications and Worksharing. 15th EAI International Conference, CollaborateCom 2019, London, UK, August 19-22, 2019, Proceedings},
        proceedings_a={COLLABORATECOM},
        year={2019},
        month={8},
        keywords={Trajectory patterns Next location prediction Self-attention Neural network},
        doi={10.1007/978-3-030-30146-0_21}
    }
    
  • Jun Zeng
    Xin He
    Haoran Tang
    Junhao Wen
    Year: 2019
    A Next Location Predicting Approach Based on a Recurrent Neural Network and Self-attention
    COLLABORATECOM
    Springer
    DOI: 10.1007/978-3-030-30146-0_21
Jun Zeng1,*, Xin He1,*, Haoran Tang1,*, Junhao Wen1,*
  • 1: Chongqing University
*Contact email: zengjun@cqu.edu.cn, hexin@cqu.edu.cn, tanghaoran@cqu.edu.cn, jhwen@cqu.edu.cn

Abstract

On most location-based social applications today, users are strongly encouraged to share activities by checking-in. In this way, vast amounts of user-generated data can be accumulated, which include spatial and temporal information. Much research has been conducted on these data, which enables heightening the understanding of human mobility. Therefore, the next location problem has attracted significant attention and has been extensively studied. In this paper, we propose a next location prediction approach based on a recurrent neural network and self-attention mechanism. Our model can explore sequence regularity and extract temporal feature according to historical trajectories information. We conduct our experiments on the location-based social network (LBSN) dataset, and the results indicate the effectiveness of our model when compared with the other three frequently-used methods.