B-MFO: A Binary Moth-Flame Optimization for Feature Selection from Medical Datasets

Advancements in medical technology have created numerous large datasets including many features. Usually, all captured features are not necessary, and there redundant irrelevant features, which reduce the performance of algorithms. To tackle this challenge, metaheuristic algorithms used to select effective However, most them scalable enough from as well small ones. Therefore, paper, a binary moth-flame optimization (B-MFO) is proposed datasets. Three categories B-MFO were developed using S-shaped, V-shaped, U-shaped transfer functions convert canonical MFO continuous binary. These evaluated on seven results compared with four well-known algorithms: BPSO, bGWO, BDA, BSSA. In addition, convergence behavior comparative assessed, statistically analyzed Friedman test. The experimental demonstrate superior solving feature selection problem for different other