Title: A novel bat algorithm fuzzy classifier approach for classification problems
Authors: Shruti Parashar; J. Senthilnath; Xin-She Yang
Addresses: Amadeus Software Labs India Pvt Ltd, Bangalore India ' Department of Aerospace Engineering, Indian Institute of Science, Bangalore, India ' School of Science and Technology, Middlesex University London, London NW4 4BT, UK
Abstract: In this paper, the application of nature-inspired algorithms (NIA) along with fuzzy classifiers is studied. The four algorithms used for the analysis are genetic algorithm, particle swarm optimisation, artificial bee colony and bat algorithm. These algorithms are used on three standard benchmark datasets and one real-time multi-spectral satellite dataset. The results obtained using different fuzzy-NIAs are analysed. Finally, we observe that the fuzzy classifiers under a given set of parameters perform more accurately when applied with the bat algorithm.
Keywords: fuzzy concepts; genetic algorithm; GA; artificial bee colony; ABC; particle swarm optimisation; PSO; bat algorithm.
International Journal of Artificial Intelligence and Soft Computing, 2017 Vol.6 No.2, pp.108 - 128
Received: 26 Dec 2015
Accepted: 24 Nov 2016
Published online: 15 Jun 2017 *