Authors: M.K. Nallakaruppan; U. Senthil Kumaran
Addresses: School of Information Technology and Engineering, VIT, Vellore, Tamil Nadu, India ' School of Information Technology and Engineering, VIT, Vellore, Tamil Nadu, India
Abstract: Swarm intelligence techniques are deployed to estimate the fitness on the search spaces and estimates the optimisation. Since the evolution of the genetic algorithm (GA) and particle swarm optimisation (PSO) optimisation problems and complex real-world problems were solved. There is a need to enhance the performance of optimisation and exploration of the search spaces. In moth-flame optimisation algorithm, the fittest moth-flame combinations with the best positions of the moth-flames after many iterations provided the optimal solutions. There is a concern for local-minima for moth-flame optimisation and the convergence rate is more, so it may skip the global optimal search. The combination of the simulated annealing (SA) and the moth-flame optimisation (MFO) provides a solution to local minima, increases the diversity of the population and increases the exploration, reduces the convergence rate to increase the performance of MFO to reach the global optima and increases the performance of MFO.
Keywords: internet of things; IoT; moth-flame optimisation; MFO; simulated annealing; SA; K-nearest neighbour classification; KNN classification; genetic algorithm; GA; particle swarm optimisation; PSO.
International Journal of Cloud Computing, 2020 Vol.9 No.2/3, pp.216 - 231
Received: 11 Apr 2019
Accepted: 11 Sep 2019
Published online: 04 Sep 2020 *