Hybrid feature selection method based on particle swarm optimization and adaptive local search method

Machine learning has been expansively examined with data classification as the most popularly researched subject. The accurateness of prediction is impacted by provided to algorithm. Meanwhile, utilizing a large amount may incur costs especially in collection and preprocessing. Studies on feature selection were mainly establish techniques that can decrease number utilized features (attributes) classification, also using generate accurate important. Hence, particle swarm optimization (PSO) algorithm suggested current article for selecting ideal set features. PSO showed be superior different domains exploring search space local algorithms are good exploiting regions. Thus, we propose hybridized an adaptive technique which works based state used accepting candidate solution. Having this combination balances intensification well global diversification searching process. surpasses original other comparable approaches, terms performance.