Bare bones particle swarm optimization with adaptive chaotic jump for feature selection in classification

Abstract

Feature selection (FS) is a crucial data pre-processing process in classification problems. It aims to reduce the dimensionality of the problem by eliminating irrelevant or redundant features while achieve similar or even higher classification accuracy than using all the features. As a variant of particle swarm optimization (PSO), Bare bones particle swarm optimization (BBPSO) is a simple but very powerful optimizer. However, it also suffers from premature convergence like other PSO algorithms, especially in high-dimensional optimization problems. In order to improve its performance in FS problems, this paper proposes a novel BBPSO based FS method called BBPSO-ACJ. An adaptive chaotic jump strategy is designed to help the stagnated particles make a large change in their searching trajectory. It can enrich the search behavior of BBPSO and prevent the particles from being trapped into local attractors. A new global best updating mechanism is employed to reduce the size of obtained feature subset. The proposed BBPSO-ACJ is compared with eight evolutionary computation (EC) based wrapper methods and two filter methods on nine benchmark datasets with different number of dimensions and instances. The experimenta l results indicate that the proposed method can select the most discriminative features from the entire feature set and achieve significantly better classification performance than other comparative methods.

A Neighborhood Based Particle Swarm Optimization with Sine Co-sine Mutation Operator for Feature Selection

“…Binary variants of grasshopper optimization algorithm (GOA) were developed to solve feature selection task in [20]. Arora et al [5] select optimal feature subset in [32] and a novel chaotic jump operator was developed to enrich the search behavior and add more randomness into the search process. Gu et al [11] introduced a new variant of PSO, called competitive swarm optimization (CSO), to solve high-dimen sional feature selection problem.…”

Self-adaptive parameter and strategy based particle swarm optimization for large-scale feature selection problems with multiple classifiers

“…PSO, which was proposed by Eberhart and Kennedy in 1995 [41], is a simple and very powerful optimizer [31]. Recently, PSO and its variations are being widely employed for solving the feature selection problems, and many PSO-based approaches have shown promising results [42,43,44,3,45]. Thus, within the framework of PSO, a new self-adaptive parameter and strategy based particle swarm optimization (SPS-PSO) algorithm is proposed in this research.…”

Variable population‐sized particle swarm optimization for highly imbalanced dataset classification

“…SD and Euclidean distance measures are used to identify the significant features and then a combination of PSO–deep belief network is used to achieve better results for classification. Qiu 21 has implemented Bare bones particle swarm optimization (BBPSO) with Adaptive Chaotic Jump (ACJ) strategy for selecting global optimal features for classification. The occurrence of premature convergence is very much reduced by using this version of PSO implementation.…”