An effective Feature Selection with Social MimicOptimization Algorithm

Hundreds of variables in data lead to data with veryhigh dimensions, allowing many feature selection methods to bedeveloped. The purpose of feature selection in machine learning,pattern recognition, and data mining is to choose features thatwill enhance learning performance. The aim of this paper is touse the binary version of the Social Mimic Optimization (SMO)algorithm as Binary Social Mimic Optimization (BSMO) forfeature selection. The combined fitness function is chosen becauseof its three main objectives: reducing classification error,balancing sensitivity and specificity, and reducing the number ofselected features. The proposed method is compared with severaloptimization methods, including Binary Genetic Algorithms(BGA) and Particle Swarm Optimization (BPSO), as well as withBinary Atom Search Optimization (BASO). The results of theevaluation using five UCI datasets show that the proposedmethod is superior to others for solving optimization problems.

کلیدواژه ها:

component ، Feature selection ، Social MimicOptimization ، Accuracy ، Sensitivity ، Specificity