地球信息科学学报 ›› 2014, Vol. 16 ›› Issue (5): 665-672.doi: 10.3724/SP.J.1047.2014.00665

所属专题: 地理大数据

• •    下一篇

大数据时代的人类移动性研究

陆锋(), 刘康, 陈洁   

  1. 中国科学院地理科学与资源研究所 资源与环境信息系统国家重点实验室,北京 100101
  • 收稿日期:2014-07-08 修回日期:2014-08-18 出版日期:2014-09-10 发布日期:2014-09-04
  • 作者简介:

    作者简介:陆 锋(1970-),博士,研究员,博士生导师。研究方向为地理信息系统理论与方法、导航与位置服务、空间数据库技术等。E-mail: luf@lreis.ac.cn

  • 基金资助:
    国家自然科学基金资助项目(41271408、41101149);国家“863”计划资助项目(2013AA120305)

Research on Human Mobility in Big Data Era

LU Feng*(), LIU Kang, CHEN Jie   

  1. State Key Lab of Resources and Environmental Information System, Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
  • Received:2014-07-08 Revised:2014-08-18 Online:2014-09-10 Published:2014-09-04
  • Contact: LU Feng E-mail:luf@lreis.ac.cn
  • About author:

    *The author: CHEN Nan, E-mail:fjcn99@163.com

摘要:

人类个体/群体移动特征是多学科共同关注的研究主题。移动定位、无线通讯和移动互联网技术的快速发展使得获取大规模、长时间序列、精细时空粒度的个体移动轨迹和相互作用定量化成为可能。同时,地理信息科学、统计物理学、复杂网络科学和计算机科学等多学科交叉也为人类移动性研究的定量化提供了有力支撑。本文首先系统总结了大数据时代开展人类移动性研究的多源异构数据基础和多学科研究方法,然后将人类移动性研究归纳为面向人和面向地理空间两大方向。面向人的研究侧重探索人类移动特性的统计规律,并建立模型解释相应的动力学机制,或分析人类活动模式,并预测出行或活动;面向地理空间的研究侧重从地理视角分析人类群体在地理空间中的移动,探索宏观活动和地理空间的交互特征。围绕这两大方向,本文评述了人类移动性的研究进展和存在问题,认为人类移动性研究在数据稀疏性、数据偏斜影响与处理、多源异构数据挖掘、机器学习方法等方面依然面临挑战,对多学科研究方法的交叉与融合提出了更高要求。

关键词: 人类移动性, 大数据, 数据挖掘, 统计物理学, 复杂网络

Abstract:

Human mobility has received much attention in many research fields such as geography, sociology, physics, epidemiology, urban planning and management in recent years. On the one hand, trajectory datasets characterized by a large scale, long time series and fine spatial-temporal granularity become more and more available with rapid development of mobile positioning, wireless communication and mobile internet technologies. On the other hand, quantitative studies of human mobility are strongly supported by interdisciplinary research among geographic information science, statistical physics, complex networks and computer science. In this paper, firstly, data sources and methods currently used in human mobility studies are systematically summarized. Then, the research is comprehended and divided into two main streams, namely people oriented and geographical space oriented. The people oriented research focuses on exploring statistical laws of human mobility, establishing models to explain the appropriate kinetic mechanism, as well as analyzing human activity patterns and predicting human travel and activities. The geographical space oriented research focuses on exploring the process of human activities in geographical space and investigating the interactions between human movement and geographical space. Followed by a detailed review of recent progress around these two streams of research, some research challenges are proposed, especially on data sparsity, data skew processing and heterogeneous data mining, indicating that more integration of multidiscipline are required in human mobility studies in the future.

Key words: human mobility, big data, data mining, statistical physics, complex network