An Improved Chicken Swarm Optimization Algorithm for Feature Selection

Abstract In recent years, feature selection is becoming more and important in data mining. Its target that reduce the dimensionality of datasets while at least maintaining classification accuracy. There are some researches about chicken swarm optimization algorithm (CSO) applied to selection, effect extraordinary compared with traditional intelligence algorithms. However, there a complex search space challenging task CSO still has default quickly gets stuck local minimum problem. An improved (ICSO) proposed this paper, which introduces Levy flight strategy hen location update nonlinear decreasing inertial weight chick increase global ability avoid getting Compared other three algorithms on eighteen UCI shows ICSO can greatly redundant features ensuring