Binary Horse Optimization Algorithm for Feature Selection

The bio-inspired research field has evolved greatly in the last few years due to large number of novel proposed algorithms and their applications. sources inspiration for these are various, ranging from behavior groups animals properties various plants. One problem is lack one algorithm which can produce best global solution all types optimization problems. presented considers proposal a approach feature selection classification problems, based on binary version algorithm. principal contributions this article are: (1) presentation main steps original Horse Optimization Algorithm (HOA), (2) adaptation HOA called Binary (BHOA), (3) application BHOA using nine state-of-the-art datasets UCI machine learning repository classifiers Random Forest (RF), Support Vector Machines (SVM), Gradient Boosted Trees (GBT), Logistic Regression (LR), K-Nearest Neighbors (K-NN), Naïve Bayes (NB), (4) comparison results with ones obtained Grey Wolf Optimizer (BGWO), Particle Swarm (BPSO), Crow Search (BCSA). experiments show that effective robust, as it returned mean accuracy value four seven datasets, respectively, compared BGWO, BPSO, BCSA, four, two, two eight, seven, five respectively.