Pattern identification using rough set clustering for spatio-temporal dataset

Spatial temporal data mining is highly demanding because of the complexity of huge amount of data collected. Interpreting the spatial data is made easier by the process of visualizing quantitative spatial data leading to the discovery of interesting patterns. To understand the spatial structure of the data, better clustering algorithms are to be used. The purpose of this paper is to investigate the efforts of applying rough set theory onto spatial clustering. Rough set theory utilizes the spatial structure of the data resulting in better pattern identification. The results of our proposed model are compared with the hard clustering methods such as single linkage, ward's and density based clustering which proves that the rough set based soft clustering converges data points faster. Our investigation shows that the rough set based clustering helps in identifying clusters that are more cohesive compared with the hard partitioning clustering method. Soft Clustering provides more information about the structure of the data than hard clustering.