Compact Coalescence Clustering Algorithm (C³A)—A GIS Anchored Approach of Clustering Discrete Points

Abstract

GIS is a subject with multi-disciplinary applications, ranging from military applications, weather forecasting, recognizing biodiversity prone regions to hotspot identification for socio-economic purposes. The main focus of the work is identification of neighboring tourist hot spots based on the Compact Coalescence Clustering Algorithm (C³A). In order to achieve this goal, firstly various tourist spots, along with the major cities and towns are identified (digitized) and based on the proposed clustering algorithm, which actually works in two phases, various existing clusters of neighborhood tourist hot spots are generated. In the first phase of the process using Clustering of Noisy Regions (CNR), several clusters of tourist spots are formed based on certain threshold distance value and accordingly the centroid is computed and updated every time the cluster is expanded. The soft clustering approach Fuzzy C-Means (FCM) is applied on the result produced in order to enhance the compactness of the clusters formed. Furthermore, the location of the tourist spots neighborhood to major cities and towns are displayed graphically on the map.