Scholarlink

This study introduces a resourceful criterion for heart failure prediction by integrating machine learning, hyper parameter tuning, and explainable artificial intelligence (XAI) techniques. Through a comparative analysis, the bagging algorithm, especially Random Forest Classifier exhibits exceptional effectiveness in accurately recognizing heart failure. The iterative hyper parameter tuning process enhances the final Random Forest Classifier model’s accuracy to an impressive 88.04%. The incorporation of XAI methods such as SHAP values and LIME, provides a subtle understanding of feature importance and individual predictions, elevating model readability. This research’s significance lies in its latent to overcome the heart failure rate. The union of machine learning classifiers, hyper parameter tuning, and XAI techniques contributes to determining whether or not a person suffers from heart failure, cultivating trust among clinicians and researchers.

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

Accurate Prediction and Classification of Heart Failure Using Machine Learning Algorithms and Interpretation Using Explainable AI

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • M. Vinith Krishna,
  • G. Rakshith,
  • Ashwini Kodipalli,
  • Trupthi Rao,
  • Jami Venkata Suman

摘要

This study introduces a resourceful criterion for heart failure prediction by integrating machine learning, hyper parameter tuning, and explainable artificial intelligence (XAI) techniques. Through a comparative analysis, the bagging algorithm, especially Random Forest Classifier exhibits exceptional effectiveness in accurately recognizing heart failure. The iterative hyper parameter tuning process enhances the final Random Forest Classifier model’s accuracy to an impressive 88.04%. The incorporation of XAI methods such as SHAP values and LIME, provides a subtle understanding of feature importance and individual predictions, elevating model readability. This research’s significance lies in its latent to overcome the heart failure rate. The union of machine learning classifiers, hyper parameter tuning, and XAI techniques contributes to determining whether or not a person suffers from heart failure, cultivating trust among clinicians and researchers.