Authors: Subrat Kumar Nayak; Pravat Kumar Rout; Alok Kumar Jagadev
Addresses: Department of Computer Science and Engineering, Siksha 'O' Anusandhan University, Bhubaneswar-30, Odisha, India ' Department of Electrical and Electronics Engineering, Siksha 'O' Anusandhan University, Bhubaneswar-30, Odisha, India ' School of Computer Engineering, KIIT University, Bhubaneswar, Odisha, India
Abstract: To arrange the uncategorised and unlabelled data into different clusters and finding the actual label of each datum from the huge volume by extracting useful and unique information is a real challenge. In this article, an automatic clustering by elitism-based multi-objective differential evolution (AC-EMODE) algorithm has been proposed to deal with partitioning the data into different clusters. This work includes three objectives to handle complex datasets. This generates a suitable Pareto front by simultaneously optimising three objectives. In addition to that, a very effective concept is followed for getting the best solution from the optimal Pareto front. A comparative analysis of the proposed approach with another six population-based methods has been carried out. These techniques are applied to ten datasets and the results reveal that the proposed approach can be considered as one of the alternative powerful methods for all data clustering applications in various fields.
Keywords: multi-objective; automatic clustering; fuzzy-based selection; differential evolution.
International Journal of Management and Decision Making, 2018 Vol.17 No.1, pp.50 - 74
Available online: 14 Dec 2017 *Full-text access for editors Access for subscribers Free access Comment on this article