Performance Comparison of Cat Swarm Optimization and Genetic Algorithm on Optimizing Functions

Abstract

This study was conducted to find out the best performance resulted from Cat Swarm Optimization (CSO) and Genetic Algorithm (GA). CSO is one of the algorithms developed based on the behavior of a number of cats to solve optimization problems. It is noted that they include two problem-solving modes such as seeking and tracing. The seeking mode happens passively and cautiously. Being cautious means become aware of the prey or other cats. The latter mode, however, occurs when the cats are active in searching the prey by moving toward it. Since they spend most of the time to idle, the seeking mode is more frequently found. In order to maximize the improvement of tracing mode, a recent calculation method was added. This study implemented three experiment functions such as sphere, Rastrigin, and knapsack. Each experiment was conducted ten times to find out the number of iterations and time needed for each method. The results show that CSO is better than GA due to its performance in terms of iterations and time.