Design and development of a Spatial DBSCAN Clustering framework for location prediction- An optimization approach

Spatial Data Mining is used in location prediction or hotspot detection in many real world applications. The objective of this work is to predict optimal cluster to place water treatment plant on the river basin using Spatial Data Mining. A multiobjective DBSCAN spatial clustering algorithm is developed to find the optimal clusters using the spatial data collected in the study area. The objective is accomplished in two phases, as the first step the data is preprocessed and mapped with the spatial features for location prediction, in the second phase the spatial data is clustered to find the optimal clusters based on the multiobjectives proposed. The results show that the locations predicted has good inward supply of water which is used in the water treatment plant based on the proposed algorithm.