Enhanced aquila optimizer for global optimization and data clustering.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Machine Learning

Background:

• The Aquila Optimizer (AO) is a metaheuristic algorithm inspired by Aquila bird behavior.
• AO exhibits limitations in high-dimensional optimization, including narrow exploration and premature local optima convergence.

Purpose of the Study:

• To introduce the Locality Opposition-Based Learning Aquila Optimizer (LOBLAO), a novel variant of AO.
• To address AO's limitations in high-dimensional optimization and improve performance in global optimization and data clustering.

Main Methods:

• Incorporation of Opposition-Based Learning (OBL) to enhance solution diversity and balance exploration/exploitation.
• Integration of a Mutation Search Strategy (MSS) to mitigate local optima and ensure robust search space exploration.

Main Results:

• LOBLAO demonstrated superior performance compared to the original AO and other state-of-the-art algorithms on benchmark functions and data clustering tasks.
• LOBLAO achieved an average rank of 1.625 in clustering problems, indicating high robustness and versatility.
• The algorithm effectively tackled high-dimensional datasets, outperforming existing methods.

Conclusions:

• LOBLAO significantly improves upon the original AO, particularly for high-dimensional optimization problems.
• The proposed enhancements (OBL and MSS) effectively address premature convergence and local optima issues.
• LOBLAO presents a robust and versatile tool for diverse and challenging optimization tasks in research and practice.