Scholarlink

Using machine learning techniques such HDBSCAN and XGBoost to cluster data points based on their qualities and classify heart disease cases as either normal or aberrant, the proposed method for the diagnosis and anomaly detection uses a reduced learning rate to reduce overfitting and regularization terms to solve some of the flaws of traditional XGBoost algorithms. The proposed method generates a more resilient clustering approach by combining the advantages of two density-based clustering algorithms, DBSCAN and OPTICS, using a hybrid density-based clustering algorithm. With a high classification accuracy of 96.58%, the experimental results of the research show that the recommended procedure outperforms current cutting-edge approaches for spotting heart disease abnormalities. More research is needed to evaluate the therapeutic value of the proposed method on larger datasets.

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

Analyzing Heart Disease Anomalies Using Hybrid Density-Based Machine Learning Technique

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Raid D. Hashim,
  • Mohammed Khaleel Jameel,
  • Mohammed Ahmed Mustafa,
  • Ekhlas A. K. Kanani,
  • Amal Mansour Hassan,
  • Adil Abbas Alwan,
  • Heba A. Abd-Alsalam Alsalame,
  • Preeti Sharma

摘要

Using machine learning techniques such HDBSCAN and XGBoost to cluster data points based on their qualities and classify heart disease cases as either normal or aberrant, the proposed method for the diagnosis and anomaly detection uses a reduced learning rate to reduce overfitting and regularization terms to solve some of the flaws of traditional XGBoost algorithms. The proposed method generates a more resilient clustering approach by combining the advantages of two density-based clustering algorithms, DBSCAN and OPTICS, using a hybrid density-based clustering algorithm. With a high classification accuracy of 96.58%, the experimental results of the research show that the recommended procedure outperforms current cutting-edge approaches for spotting heart disease abnormalities. More research is needed to evaluate the therapeutic value of the proposed method on larger datasets.