Chaotic Binary Particle Swarm Optimization for Feature Selection using Logistic Map
Li‐Yeh Chuang, Jung-Chike Li, Cheng‐Hong Yang · 2008
Abstract—Feature selection is a useful technique for increasing classification accuracy. The primary objective is to remove irrelevant features in the feature space and identify relevant features. Binary particle swarm optimization (BPSO) has been applied successfully in solving feature selection problem. In this paper, chaotic binary particle swarm optimization (CBPSO) with logistic map for determining the inertia weight is used. The K-nearest neighbor (K-NN) method with leave-one-out cross-validation (LOOCV) serves as a classifier for evaluating classification accuracies. Experimental results indicate that the proposed method not only reduces the number of features, but also achieves higher classification accuracy than other methods. Index Terms—Feature selection, binary particle swarm optimization, logistic map, K-nearest neighbor, leave-one-out cross-validation. I.