Prediction of hepatitis C-related liver diseases through ensemble learning: a comprehensive analysis using the UCI HCV dataset

Hepatitis C Virus (HCV) remains a significant global health concern, affecting an estimated 50 million individuals worldwide, with nearly 1 million new cases reported annually, according to the World Health Organization. Early detection and accurate classification of liver complications related to HCV are essential for timely and effective clinical intervention. This study explores the HCV dataset from the UCI Machine Learning Repository to evaluate the predictive performance of three ensemble learning algorithms: Random Forest, AdaBoost, and XGBoost. Comprehensive pre-processing steps, including data visualization, normalization, and class balancing using ADASYN, were applied. Comparative analysis revealed that while XGBoost performed best on the raw imbalanced data, Random Forest achieved the highest overall accuracy (0.99) after applying ADASYN. The findings underscore the potential of ensemble learning methods, particularly when combined with appropriate data balancing techniques, to support early diagnosis and clinical decision-making in liver disease management.