An Improved Particle Swarm Optimization Algorithm for Data Classification

Optimisation-based methods are enormously used in the field of data classification. Particle Swarm Optimization (PSO) is a metaheuristic algorithm based on swarm intelligence, widely to solve global optimisation problems throughout real world. The main problem PSO faces premature convergence due lack diversity, and it usually stuck local minima when dealing with complex real-world problems. In meta-heuristic algorithms, population initialisation an important factor affecting diversity speed. this study, we propose improved variant that enhances speed by applying pseudo-random sequences opposite rank inertia weights instead using random distributions for initialisation. This paper also presents novel method quasi-random sequence (Faure) create swarm, through opposition-based method, generated. We proposed opposition rank-based weight approach adjust particles increase performance standard PSO. (ORIW-PSO-F) has been tested optimise feed-forward neural network fifteen sets taken from UCI. techniques’ experiment result depicts much better than other existing techniques.