Abstract:Feature selection, an optimization problem, becomes an important pre-process tool in data mining, which simultaneously aims at minimizing feature-size and maximizing model generalization. Because of large search space, conventional optimization methods often fail to generate global optimum solution. A variety of hybrid techniques merging different search strategies have been proposed in feature selection literature, but mostly deal with low dimensional datasets. In this paper, a hybrid optimization method is proposed for numerical optimization and feature selection, which integrates sine-cosine algorithm (SCA) in Harris hawks optimization (HHO). The goal of SCA integration is to cater ineffective exploration in HHO, moreover exploitation is enhanced by dynamically adjusting candidate solutions for avoiding solution stagnancy in HHO. The proposed method, namely SCHHO, is evaluated by employing CEC?17 test suite for numerical optimization and sixteen datasets with low and high-dimensions exceeding 15000 attributes, and compared with original SCA and HHO, as well as, other well-known optimization methods like dragonfly algorithm (DA), whale optimization algorithm (WOA), grasshopper optimization algorithm (GOA), Grey wolf optimization (GWO), and salp swarm algorithm (SSA); in addition to state-of-the-art methods. Performance of the proposed method is also validated against hybrid methods proposed in recent related literature. The extensive experimental and statistical analyses suggest that the proposed hybrid variant of HHO is able to produce efficient search results without additional computational cost. With increased convergence speed, SCHHO reduced feature-size up to 87% and achieved accuracy up to 92%. Motivated from the findings of this study, various potential future directions are also highlighted.

Spotted Hyena Optimization Algorithm with Simulated Annealing for Feature Selection

Hybrid Whale Optimization Algorithm with simulated annealing for feature selection

A New Hybrid Seagull Optimization Algorithm for Feature Selection.

A Hybrid Approach of Spotted Hyena Optimization Integrated with Quadratic Approximation for Training Wavelet Neural Network

Advances in Spotted Hyena Optimizer: A Comprehensive Survey

A hybrid Harris Hawks optimization algorithm with simulated annealing for feature selection

Enhancing Spotted Hyena optimization with fuzzy logic for complex engineering optimization

Using spotted hyena optimizer for training feedforward neural networks

An Efficient Hybrid Sine-Cosine Harris Hawks Optimization For Low And High-Dimensional Feature Selection

Multi-objective spotted hyena optimizer: A Multi-objective optimization algorithm for engineering problems

Cross-Scene Hyperspectral Feature Selection Via Hybrid Whale Optimization Algorithm with Simulated Annealing

In Search of Excellence: SHOA as a Competitive Shrike Optimization Algorithm for Multimodal Problems

Feature selection based on dataset variance optimization using Hybrid Sine Cosine – Firehawk Algorithm (HSCFHA)

Feature Selection Based on Improved White Shark Optimizer

Boosted Binary Harris Hawks Optimizer and Feature Selection

Evolutionary Feature Selection Based on Hybrid Bald Eagle Search and Particle Swarm Optimization

Novel Meta-Heuristic Algorithm for Feature Selection, Unconstrained Functions and Engineering Problems

S-Shaped vs. V-Shaped Transfer Functions for Antlion Optimization Algorithm in Feature Selection Problems

Multi-objective particle swarm optimization with adaptive strategies for feature selection

A Novel Snow Leopard Optimization for High-Dimensional Feature Selection Problems