Orthogonal Learning Rosenbrock’s Direct Rotation with the Gazelle Optimization Algorithm for Global Optimization

Abstract

An efficient optimization method is needed to address complicated problems and find optimal solutions. The gazelle optimization algorithm (GOA) is a global stochastic optimizer that is straightforward to comprehend and has powerful search capabilities. Nevertheless, the GOA is unsuitable for addressing multimodal, hybrid functions, and data mining problems. Therefore, the current paper proposes the orthogonal learning (OL) method with Rosenbrock’s direct rotation strategy to improve the GOA and sustain the solution variety (IGOA). We performed comprehensive experiments based on various functions, including 23 classical and IEEE CEC2017 problems. Moreover, eight data clustering problems taken from the UCI repository were tested to verify the proposed method’s performance further. The IGOA was compared with several other proposed meta-heuristic algorithms. Moreover, the Wilcoxon signed-rank test further assessed the experimental results to conduct more systematic data analyses. The IGOA surpassed other comparative optimizers in terms of convergence speed and precision. The empirical results show that the proposed IGOA achieved better outcomes than the basic GOA and other state-of-the-art methods and performed better in terms of solution quality.