New Modified Controlled Bat Algorithm for Numerical Optimization Problem

Bat algorithm (BA) is an eminent meta-heuristic that has been widely used to solve diverse kinds of optimization problems. BA leverages the echolocation feature bats produced by imitating bats’ searching behavior. faces premature convergence due its local search capability. Instead using standard uniform walk, Torus walk viewed as a promising alternative improve In this work, we proposed improved variation applying torus diversity and convergence. The proposed. Modern Computerized Algorithm (MCBA) approach examined for fifteen well-known benchmark test finding our technique shows performance compared PSO BA. MCBA, BPA, Standard PSO, have problems training a.pngicial neural network (ANN). We performed experiments eight datasets applied from worldwide famous machine-learning (ML) repository UCI. Simulation results shown ANN with MCBA-NN tops list considering exactness, more superiority traditional methodologies. may be effectively data classification statistical in future.