期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2014
卷号：14
期号：1
页码：72-76
出版社：International Journal of Computer Science and Network Security
摘要：In order to verify the clustering results, domain experts usually require it to be represented in interpretable and meaningful ways. For that reason, the spatio-temporal clusters obtained are essentially needed to be visualized in an understandable view to support visual exploration of cluster structures. Scatter-Principal Component Analysis (Scatter-PCA) is proposed to visualize the spatio-temporal clustering result. Scatter-PCA combines PCA that projected m-dimensional spatio-temporal data into 2 dimensional spatio-temporal data with scatter plot to visualize the structure of clusters. Two spatio-temporal data: crime data and traffic accident data are utilized to validate the visual clustering approach. The experimental results on two clustering result of spatio-temporal data demonstrate the effectiveness of our visual clustering approach to investigate the structure of clusters.
关键词：Cluster; visualization; interpretation; principal component analysis; scatter plot
