《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (7): 2100-2106.DOI: 10.11772/j.issn.1001-9081.2022091364

• 第39届CCF中国数据库学术会议(NDBC 2022) • 上一篇    

融合出发地与目的地时空相关性的城市区域间出租车需求预测

魏远1,2, 林彦1,2, 郭晟楠1,2,3, 林友芳1,2,3, 万怀宇1,2,3()   

  1. 1.北京交通大学 计算机与信息技术学院, 北京 100044
    2.交通数据分析与挖掘北京市重点实验室(北京交通大学), 北京 100044
    3.民航旅客服务智能化应用技术重点实验室, 北京 100044
  • 收稿日期:2022-07-12 修回日期:2022-09-19 接受日期:2022-10-12 发布日期:2023-07-20 出版日期:2023-07-10
  • 通讯作者: 万怀宇
  • 作者简介:魏远(1997—),男,河南许昌人,硕士研究生,CCF会员,主要研究方向:深度学习、时空数据挖掘;
    林彦(1998—),男,江西新余人,博士研究生,主要研究方向:深度学习、时空数据挖掘;
    郭晟楠(1992—),女,辽宁沈阳人,讲师,博士,CCF会员,主要研究方向:时空数据挖掘、智能交通;
    林友芳(1971—),男,福建龙岩人,教授,博士,CCF高级会员,主要研究方向:数据挖掘、机器学习、强化学习;
    万怀宇(1981—),男,湖北宣恩人,副教授,博士,CCF会员,主要研究方向:时空数据挖掘、信息抽取、社交网络挖掘。
  • 基金资助:
    中国博士后科学基金资助项目(2021M700365)

Prediction of taxi demands between urban regions by fusing origin-destination spatial-temporal correlation

Yuan WEI1,2, Yan LIN1,2, Shengnan GUO1,2,3, Youfang LIN1,2,3, Huaiyu WAN1,2,3()   

  1. 1.School of Computer and Information Technology,Beijing Jiaotong University,Beijing 100044,China
    2.Beijing Key Laboratory of Traffic Data Analysis and Mining (Beijing Jiaotong University),Beijing 100044,China
    3.CAAC Key Laboratory of Intelligent Passenger Service of Civil Aviation,Beijing 100044,China.
  • Received:2022-07-12 Revised:2022-09-19 Accepted:2022-10-12 Online:2023-07-20 Published:2023-07-10
  • Contact: Huaiyu WAN
  • About author:WEI Yuan, born in 1997, M. S. candidate. His research interests include deep learning, spatial-temporal data mining.
    LIN Yan, born in 1998, Ph. D. candidate. His research interests include deep learning, spatial-temporal data mining.
    GUO Shengnan, born in 1992, Ph. D., lecturer. Her research interests include spatial-temporal data mining, intelligent transportation.
    LIN Youfang, born in 1971, Ph. D., professor. His research interests include data mining, machine learning, reinforcement learning.
    WAN Huaiyu, born in 1981, Ph. D., associate professor. His research interests include spatial-temporal data mining, information extraction, social network mining.
  • Supported by:
    China Postdoctoral Science Foundation(2021M700365)

摘要:

精准预测城市区域之间的出租车需求量,可以为出租车的引导和调度以及乘客的出行推荐提供决策支持信息,从而优化出租车的供需关系。然而现有模型大多以区域内的出租车需求量为建模和预测对象,对区域之间的时空相关性考虑不足,且较少关注区域之间更细粒度的需求量预测。针对上述问题,提出一种面向城市区域间出租车需求量的预测模型——出发地—目的地融合时空网络(ODSTN)模型。该模型分别从区域和区域对两个空间维度以及临近、日、周三个时间维度出发,采用图卷积和时间注意力机制来捕获区域之间的复杂时空相关性,并设计了一种新的路径感知融合机制来对多角度的特征进行融合,最终实现了对城市区域间出租车需求量的预测。在成都和曼哈顿地区两个真实出租车订单数据集中进行了实验,结果表明ODSTN模型的平均绝对误差(MAE)、均方根误差(RMSE)和平均绝对百分比误差(MAPE)分别为0.897 1、3.527 4、50.655 6%和0.589 6、1.163 8、61.079 4%。可见,ODSTN模型在出租车需求预测任务上具有较高的准确性。

关键词: 出租车需求预测, 时空相关性, 出发地—目的地, 图卷积网络, 注意力机制

Abstract:

Accurate prediction of taxi demands between urban regions can provide decision support information for taxi guidance and scheduling as well as passenger travel recommendation, so as to optimize the relation between taxi supply and demand. However, most of the existing models only focus on modeling and predicting the taxi demands within a region, do not consider enough the spatial-temporal correlation between regions, and pay less attention to the more fine-grained demand prediction between regions. To solve the above problems, a prediction model for taxi demands between urban regions — Origin-Destination fusion with Spatial-Temporal Network (ODSTN) model was proposed. In this model, complex spatial-temporal correlations between regions was captured from spatial dimensions of the regions and region pairs respectively and three temporal dimensions of recent, daily and weekly periods by using graph convolution and attention mechanism, and a new path perception fusion mechanism was designed to combine the multi-angle features and finally realize the taxi demand prediction between urban regions. Experiments were carried out on two real taxi order datasets in Chengdu and Manhattan. The results show that the Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) of ODSTN model are 0.897 1, 3.527 4, 50.655 6% and 0.589 6, 1.163 8, 61.079 4%, respectively, indicating that ODSTN model has high accuracy in taxi demand prediction tasks.

Key words: taxi demand prediction, spatial-temporal correlation, origin-destination, Graph Convolutional Network (GCN), attention mechanism

中图分类号: