Scholarlink

Hepatitis C Virus (HCV) continues to pose a vital global health challenge, affecting around 50 million individuals worldwide, with approximately 1 million new cases reported yearly, according to the World Health Organization. Early detection and precise classification of HCV-related liver complications are critical for timely and effective clinical intervention. This study utilizes the HCV dataset from the UCI Machine Learning Repository to assess the predictive capabilities of three ensemble learning algorithms: Random Forest, AdaBoost, and XGBoost. The data underwent comprehensive preprocessing, including visualization, normalization, and class balancing using the ADASYN technique. Comparative analysis demonstrated that while XGBoost exhibited superior performance on the original imbalanced dataset, Random Forest achieved the highest overall accuracy (0.99) following ADASYN application. These results highlight the efficacy of ensemble learning approaches, especially when paired with appropriate data balancing methods, in facilitating early diagnosis and supporting clinical decision-making in the management of liver diseases.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Hepatitis C-Related Liver Diseases Through Ensemble Learning: A Comprehensive Analysis Using the UCI HCV Dataset

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Md Fahim Hossain,
  • Khurshida Jahan Moumi,
  • Maitreyee Dey

摘要

Hepatitis C Virus (HCV) continues to pose a vital global health challenge, affecting around 50 million individuals worldwide, with approximately 1 million new cases reported yearly, according to the World Health Organization. Early detection and precise classification of HCV-related liver complications are critical for timely and effective clinical intervention. This study utilizes the HCV dataset from the UCI Machine Learning Repository to assess the predictive capabilities of three ensemble learning algorithms: Random Forest, AdaBoost, and XGBoost. The data underwent comprehensive preprocessing, including visualization, normalization, and class balancing using the ADASYN technique. Comparative analysis demonstrated that while XGBoost exhibited superior performance on the original imbalanced dataset, Random Forest achieved the highest overall accuracy (0.99) following ADASYN application. These results highlight the efficacy of ensemble learning approaches, especially when paired with appropriate data balancing methods, in facilitating early diagnosis and supporting clinical decision-making in the management of liver diseases.