Abstract—Nature-inspired algorithms have been successful for more efficient clustering of unlabeled data, and have effectively been used to improve a wide variety of numerical optimization problems, and when these algorithms are combined with suitable objective functions, the centroid for clusters is determined iteratively. Centroids are the points that are closest to the center in a cluster. These algorithms are not without their shortcomings, such as slow convergence or fixating on local minima. These are just some of the minor inconveniences that might be caused during our procedures of trying to create a hybrid. A recent trend that has been observed is the hybridization of these algorithms to overcome the shortcomings of the vanilla versions of the algorithm for efficient optimization and clustering. In this work, a novel version of such a Hybrid Meta-heuristic algorithm, developed from the Firefly and Whale optimization algorithms, for faster convergence and better optimization compared to its vanilla counterparts is presented. The firefly and whale algorithms are hybridized such that the drawbacks of one algorithm are taken care of and compensated by the advantages of the other. The outcomes show that the hybrid algorithm of whale and firefly converges faster and is more efficient in comparison with other nature inspired algorithms and its efficiency is further established from the results on standard datasets and as well for finding the clusters with in the different subspaces.
 
Keywords—subspace clustering algorithm, nature inspired algorithm, hybrid meta-heuristic algorithm, efficient clustering, firefly algorithm, whale algorithms


Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.