Boosting Atomic Orbit Search Using Dynamic-Based Learning for Feature Selection

Feature selection (FS) is a well-known preprocess step in soft computing and machine learning algorithms. It plays critical role different real-world applications since it aims to determine the relevant features remove other ones. This process (i.e., FS) reduces time space complexity of technique used handle collected data. The feature methods based on metaheuristic (MH) techniques established their performance over all conventional FS methods. So, this paper, we presented modified version new MH named Atomic Orbital Search (AOS) as technique. performed using advances dynamic opposite-based (DOL) strategy that enhance ability AOS explore search domain. by increasing diversity solutions during searching updating A set eighteen datasets has been evaluate efficiency developed approach, AOSD, results AOSD are compared with From results, can reduce number preserving or classification accuracy better than techniques.