Metaheuristic Machine Learning Algorithms for Liver Disease Prediction

In machine learning, optimizing solutions is critical for improving performance. This study explores the use of metaheuristic algorithms to enhance key processes such as hyperparameter tuning, feature selection, and model optimization. Specifically, we integrate the Artificial Bee Colony (ABC) algorithm with Random Forest and Decision Tree models to improve the accuracy and efficiency of disease prediction. Machine learning has the potential to uncover complex patterns in medical data, offering transformative capabilities in disease diagnosis. However, selecting the optimal algorithm for model optimization presents a significant challenge. In this work, we employ Random Forest, Decision Tree models, and the ABC algorithm—based on the foraging behaviours of honeybees—to predict liver disease using a dataset from Indian medical records. Our experiments demonstrate that the Random Forest model achieves an accuracy of 85.12%, the Decision Tree model 76.89%, and the ABC algorithm 80.45%. These findings underscore the promise of metaheuristic approaches in machine learning, with the ABC algorithm proving to be a valuable tool in improving predictive accuracy. In conclusion, the integration of machine learning models with metaheuristic techniques, such as the ABC algorithm, represents a significant advancement in disease prediction, driving progress in data-driven healthcare.