Clustering Assisted Co-location Pattern Mining for Spatial Data

The importance of spatial data mining is growing with the increasing incidence and importance of large spatial datasets repositories of remote-sensing images, location based mobile app data, satellite imagery, medical data and crime data with location information, three dimensional maps, traffic data and many more. However, as classical data mining techniques are often inadequate for spatial data mining, different techniques for spatial data mining are being exclusively developed. Colocation pattern mining is one of the techniques to discover the set of spatial features frequently located together in the geographic proximity. In this paper we propose a model for finding the frequently occurring co-location patterns of objects in spatial datasets using a co-location mining algorithm which utilizes clustering technique before mining the data for rules. The proposed algorithm namely, SigCPM overcomes the limitations of the existing grid-based approaches. The proposed algorithm is compared with the existing methods of co-location pattern mining and evaluated.