Boosted Sine Cosine Algorithm with Application to Medical Diagnosis

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Machine Learning

Background:

• The Sine Cosine Algorithm (SCA) is effective for optimization but suffers from low population diversity and local optima stagnation.
• Hybridization of optimization algorithms is a common strategy to overcome individual algorithm limitations.

Purpose of the Study:

• To propose an enhanced Sine Cosine Algorithm with Particle Swarm Optimization (ESCA_PSO).
• To improve population diversity and avoid local optima in global optimization tasks.
• To evaluate the performance of ESCA_PSO on benchmark functions and medical diagnosis tasks.

Main Methods:

• Hybridization of the Sine Cosine Algorithm (SCA) with Particle Swarm Optimization (PSO).
• Incorporation of multiple mutation strategies into the hybrid algorithm.
• Validation using diverse benchmark functions and Support Vector Machine (SVM) parameter tuning for medical diagnosis.

Main Results:

• ESCA_PSO demonstrated superior performance compared to other algorithms on various benchmark functions.
• The proposed algorithm effectively addressed issues of low population diversity and local optima.
• ESCA_PSO successfully optimized Support Vector Machine parameters for medical diagnosis tasks, proving its practical applicability.

Conclusions:

• ESCA_PSO is an effective hybrid optimization algorithm that overcomes limitations of the original SCA.
• The enhanced algorithm shows significant potential for solving complex global optimization problems.
• ESCA_PSO offers a robust approach for parameter tuning in machine learning applications like medical diagnosis.