Spectral Clustering based Approach on Large Water Distribution Network

Abstract: Minimum spanning tree(MST) based clustering approach is a powerful tool for performing cluster analysis on large graphs. The algorithm produces k-clusters with the MST. To identify the crucial components, MST based clustering approach plays a major role. Real world applications like Water Distribution Networks are complex networks that require innovative technological solutions for efficient management of the system. Spectral clustering algorithm is more effective than traditional clustering algorithm. In this paper we propose a clustering approach on Water Distribution Network. We suggest to find the MST for the large graph and apply spectral clustering algorithm to divide the MST into K-clusters. Spectral clustering algorithm is capable of detecting clusters with irregular boundaries. Experimental results show that the proposed approach performed well and found to fit to the expected results for both synthetic data and real world data.

Keywords: Large graph, Minimum Spanning Tree, Clustering, Spectral Clustering, Water distribution network.