A Clustering Scheme for Trajectories in Road Networks
Trajectories of moving objects contain numerous information to analyze and exploit Clustering analysis is an important approach to deal with trajectories in data mining technology. However, existing trajectory clustering algorithm barely considers temporal information during clustering procedure. This paper proposes a clustering scheme for trajectories in road network environment. A trajectory on road network is represented by a sequence of interest points that each point indicates a real location on road segment. We define a distance measure to compute the spatial similarity between trajectories, then we propose a clustering algorithm based on DBSCAN and add temporal factor into it at the same time.
trajectory clustering distance measure density-based method
Yalin Wang Qilong Han Haiwei Pan
College of Computer Science and Technology Harbin Engineering University Harbin, China
国际会议
北京
英文
373-376
2010-09-18(万方平台首次上网日期,不代表论文的发表时间)