Comparison of Binary Particle Swarm Optimization And Binary Dragonfly Algorithm for Choosing the Feature Selection

Abstract

Artificial life is a living behavior that comes from animals and humans which is currently used as inspiration in making algorithms and is usually used to look for patterns such as classification, feature selection, optimization, and many more. Binary Particle Swarm Optimization (BPSO) and Binary Dragonfly Algorithm (BDA) are algorithms derived from Artificial Life and Genetic Algorithms. The purpose of this research is to compare Binary Particle Swarm Optimization (BPSO), Binary Dragonfly Algorithm (BDA). The dataset used is breast cancer from the UCI public dataset. The results of the feature selection will be used in the K-NN (K Nearest Neighbor) classification algorithm to get the Accuracy, Precision, Recall values. The BPSO algorithm is still superior to the BDA algorithm because the iteration process of BPSO algorithm is more centralized in solving existing problems. the BPSO algorithm produces 97.5% accuracy, 97% precision, and 98.98% recall compared to the BDA algorithm whose value is still below BPSO.