A Delaunay Triangulation Preprocessing Based Fuzzy-Encroachment Graph Clustering for Large Scale GIS Data

Abstract

This paper proposed a time efficient graph based spatial clustering for large scale GIS data. As volume of GIS data is large, the direct clustering will not be that much efficient in both space and time complexity domains. So, data is preprocessed using Delaunay Triangulation to reduce both the space and time complexities. The preprocessed data is then considered for spanning tree based crisp clustering. The Fuzzy based postprocessing refinement is used to incorporate some extra points. The time and space complexity has been reduced and as a result efficiency of clustering is achieved.