Hybrid framework for DBSCAN algorithm using fuzzy logic

Data mining process is to obtain information from a data set and then convert it into an understandable and meaningful information for further use. DBSCAN, a density based clustering algorithm, identifies clusters of varying shape and outliers. DBSCAN is based on bivalent logic. Therefore it can only detect objects as completely belonging to a particular cluster or not wholly belonging to it. In this paper, a framework of methodology of DBSCAN algorithm with the integration of fuzzy logic is proposed. The extent to which an object belongs to a particular cluster will be determined using membership values. The improved version of DBSCAN algorithm will be the hybridization of DBSCAN algorithm with fuzzy if-then rules.