Teaching Learning-Based Optimization With Evolutionary Binarization Schemes for Tackling Feature Selection Problems

Machine learning techniques heavily rely on available training data in a set. Certain features the can interfere with process, so it is required to remove irrelevant and redundant build robust model. As such, several feature selection are usually applied pre-processing phase obtain most appropriate set of improve overall process. In this paper, new approach proposed based modified Teaching-Learning-based Optimization (TLBO) combined four binarization methods: Elitist, Elitist Roulette, Tournament, Rank-based method. The influence these methods studied compared other state-of-the-art techniques. experimental results such as Shapiro-Wilk normality Wilcoxon ranksum test show that both transfer functions approaches have significant effectiveness binary TLBO. experiments choosing fitting function along suitable method has substantial impact exploratory exploitative potentials technique.