Hybrid Swarm Intelligence Optimization of Three Machine Learning Classifiers for Chronic Kidney Disease Detection

Effective and reliable classification methods are crucial for the timely diagnosis and management of Chronic Kidney Disease (CKD). This research introduces a Hybrid Swarm Intelligence Optimization approach that combines the Firefly Algorithm (FA) and Grey Wolf Optimizer (GWO), aimed at enhancing the performance of three widely used machine learning classifiers: Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbors (KNN). The study evaluated classifier performance using the CKD dataset and an additional highly imbalanced dataset, reflecting realistic clinical scenarios. Results demonstrate that the hybrid FA-GWO approach substantially improved classifier performance, notably achieving perfect accuracy and AUC scores with the Random Forest classifier on the CKD dataset, while effectively managing class imbalance. The findings underscore the practical potential of hybrid optimization techniques for clinical diagnostic applications, providing a valuable benchmark for future research. Detailed step-by-step procedures, pseudo-code, and full hyperparameter settings supporting this study are provided in a supplementary document available at: https://github.com/NarongdechGithub/Hybrid-Swarm-Intelligence.git