Title :
Hybrid framework for DBSCAN algorithm using fuzzy logic
Author :
Beri, Saefia ; Kaur, Kamaljit
Author_Institution :
Dept. of Comput. Sci. & Eng., Guru Nanak Dev Univ., Amritsar, India
Abstract :
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.
Keywords :
data mining; fuzzy logic; fuzzy reasoning; pattern clustering; DBSCAN algorithm; bivalent logic; data mining process; density based clustering algorithm; density-based spatial clustering-of-application-with-noise; fuzzy if-then rules; fuzzy logic; hybrid framework; membership values; Algorithm design and analysis; Breast cancer; Classification algorithms; Clustering algorithms; Data mining; Noise; Spatial databases; DBSCAN; bivalent logic; clustering;
Conference_Titel :
Futuristic Trends on Computational Analysis and Knowledge Management (ABLAZE), 2015 International Conference on
Conference_Location :
Noida
Print_ISBN :
978-1-4799-8432-9
DOI :
10.1109/ABLAZE.2015.7155024
