Title :
A New Trajectory Clustering Based on Partition-Cluster-Extraction
Author :
Jiashun Chen ; Dechang Pi
Author_Institution :
Coll. of Comput. Sci. & Technol., HHIT, Lianyungang, China
Abstract :
In this paper, we proposed a new trajectory clustering based on partition-cluster-extract (PCE). Firstly, some relative definitions are defined, and then we proposed a new partition method named PCE, which is based on method of partition-group framework. Finally, on the basis of PCE and relative definitions, we proposed a new algorithm named partition-cluster-extract trajectory clustering (PCETC). Some experiments on different datsets show that the new algorithm in this paper is more efficient than TRACLUS, at the same time, the algorithm has nearly the same cluster quality as TRACLUS.
Keywords :
data analysis; pattern clustering; PCE; TRACLUS; cluster quality; partition-cluster-extract trajectory clustering; partition-group framework; Algorithm design and analysis; Classification algorithms; Clustering algorithms; Joints; Partitioning algorithms; Spatiotemporal phenomena; Trajectory; cluster; partition-cluster-extract; trajectory;
Conference_Titel :
Computational and Information Sciences (ICCIS), 2013 Fifth International Conference on
Conference_Location :
Shiyang
DOI :
10.1109/ICCIS.2013.187