DATA MINING FOR PARAMETER SELECTION OF SWARM INTELLIGENCE ALGORITHMS

Swarm Intelligence algorithms commonly used to solve optimization problems. This study considers the problem of the parameters selection of the Particle Swarm Optimization algorithm. Methods of data mining are proposed to use for the selection. An example of applying regression analysis and classifying for Particle Swarm Optimization are given. The analysis carried out allows us to find good parameters of the Particle Swarm Optimization algorithm for a test optimization problem. The effectiveness of parameters found has been compared with parameters recommended by other researchers.