Scholarlink

Cardiovascular disease ranks among the top causes of mortality globally, highlighting the urgent need for the formulation of accurate and efficient predictive models. Machine learning algorithms assist in making it possible to recognize cardiac problems easily and at a reasonable cost. This research utilizes the CHSLB (Cleveland, Hungary, Switzerland, and Long Beach) and Cleveland records to evaluate the results of various machine learning approaches for predicting cardiovascular disease, including Support Vector Machine, Random Forest, AdaBoost, Decision Tree, K Nearest Neighbors, XGBoost, Naïve Bayes, Logistic Regression, Bagging, and Voting classifiers. For the CHSLB dataset, the pinnacle of accuracy is reached by Random Forest, Decision Tree, XGBoost, and Bagging at 99.03%. Conversely, the Cleveland dataset displayed differing performance patterns, with Naïve Bayes and Logistic Regression obtaining the best accuracy of 88.63%. Ensemble algorithms, particularly Random Forest, Bagging, and XGBoost, showed less success on Cleveland data, underscoring their reliability in heart disease forecasting. This study informs the clinical decision process by providing insight regarding the selection of appropriate machine-learning strategies for heart disease recognition.

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

Heart Disease Prediction Using Machine Learning: A Comparative Approach on CHSLB and Cleveland Datasets with Advanced Modeling Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Karanpreet Kumar,
  • Roopali Garg,
  • Neelam Goel

摘要

Cardiovascular disease ranks among the top causes of mortality globally, highlighting the urgent need for the formulation of accurate and efficient predictive models. Machine learning algorithms assist in making it possible to recognize cardiac problems easily and at a reasonable cost. This research utilizes the CHSLB (Cleveland, Hungary, Switzerland, and Long Beach) and Cleveland records to evaluate the results of various machine learning approaches for predicting cardiovascular disease, including Support Vector Machine, Random Forest, AdaBoost, Decision Tree, K Nearest Neighbors, XGBoost, Naïve Bayes, Logistic Regression, Bagging, and Voting classifiers. For the CHSLB dataset, the pinnacle of accuracy is reached by Random Forest, Decision Tree, XGBoost, and Bagging at 99.03%. Conversely, the Cleveland dataset displayed differing performance patterns, with Naïve Bayes and Logistic Regression obtaining the best accuracy of 88.63%. Ensemble algorithms, particularly Random Forest, Bagging, and XGBoost, showed less success on Cleveland data, underscoring their reliability in heart disease forecasting. This study informs the clinical decision process by providing insight regarding the selection of appropriate machine-learning strategies for heart disease recognition.