A state-adaptive booby optimization algorithm for engineering design and medical data applications.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• Balancing global exploration and local exploitation is a key challenge in metaheuristic optimization.
• High-dimensional, nonlinear, and constrained problems are prevalent in engineering and medical data analysis.

Purpose of the Study:

• To introduce the Booby Optimization Algorithm (BOA), a novel state-adaptive metaheuristic.
• To evaluate BOA's effectiveness on benchmark functions, engineering design problems, and medical data analysis tasks.

Main Methods:

• BOA utilizes an adaptive state variable to control search strategies and transitions.
• The algorithm incorporates nonlinear dynamics, stochastic perturbations, and diversity preservation mechanisms.
• BOA was tested on CEC benchmark functions, engineering design problems, and 14 real-world medical datasets.

Main Results:

• BOA consistently outperformed state-of-the-art metaheuristics in convergence speed, solution accuracy, and robustness.
• Near-optimal or best-known solutions were achieved with reduced mean error and variance.
• In medical classification, BOA-based feature selection achieved 96.20% mean accuracy with high sensitivity and specificity.

Conclusions:

• BOA is an efficient and robust adaptive optimization framework.
• The algorithm is well-suited for complex engineering optimization and medical decision-support applications.
• Rigorous statistical validation confirmed BOA's significant improvements.