Binary Growth Optimizer: For Solving Feature Selection Optimization Problems

The present study introduces a novel binary meta-heuristic optimizer, referred to as the Binary Growth Optimizer (BGO), which is specifically developed to address discrete optimization problems. The BGO utilizes transfer functions to convert positions in a continuous space to discrete positions. To assess the performance of the proposed algorithm, we conducted experiments on 14 publicly available datasets from the UCI Machine Learning Repository. We compared the BGO’s performance with seven state-of-the-art meta-heuristics. Based on convergence accuracy statistics, the results indicate that the BGO outperforms the seven compared meta-heuristics, providing the most favorable outcomes.