Scholarlink

In the medical field, predict cardiac disease is thought to be one of the hardest tasks. Researchers used a number of methods, such as LDA, RF, GBC, DT, SVM, and KNN, in count to the sequence choice of pattern assortment method, to find cardiac infection. For identification, the system employs the cross-validation K-fold technique. These six approaches were used in the contrasting study. Along with the Heart Statlog Cleveland Hungary dataset, the predictive efficacy was assessed employing the respective data sets comprising Cleveland, Hungray, Switzerland, as well as Long Beach V. In both of the Hungary, Switzerland as well as Long Beach V along with Heart Statlog Cleveland Hungary Datasets, Random Forest Classifier sfs as well as Decision Tree Classifier sfs achieved the highest and almost equivalent precision ratings (100%, 99.40% and 100%, 99.76%, correspondingly). The results were contrasted with earlier studies on cardiac prediction. We intend to develop the model further in the future so that it might be applied to different feature selection strategies. Alternatively, a random forest classifier could be employed. The most important objective of this study is to enhance earlier effort by devising a novel and unique model-creation method and to render the model applicable and user-friendly in practical contexts.

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

Employing Machine Learning Algorithms for Cardiac Illness Predictability Using Feature Selection Method

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • K. Srujan Raju,
  • Banothu Ramji,
  • Rajesh Tiwari,
  • Sheo Kumar,
  • V. A. Narayana,
  • Manyala Naga Sailaja

摘要

In the medical field, predict cardiac disease is thought to be one of the hardest tasks. Researchers used a number of methods, such as LDA, RF, GBC, DT, SVM, and KNN, in count to the sequence choice of pattern assortment method, to find cardiac infection. For identification, the system employs the cross-validation K-fold technique. These six approaches were used in the contrasting study. Along with the Heart Statlog Cleveland Hungary dataset, the predictive efficacy was assessed employing the respective data sets comprising Cleveland, Hungray, Switzerland, as well as Long Beach V. In both of the Hungary, Switzerland as well as Long Beach V along with Heart Statlog Cleveland Hungary Datasets, Random Forest Classifier sfs as well as Decision Tree Classifier sfs achieved the highest and almost equivalent precision ratings (100%, 99.40% and 100%, 99.76%, correspondingly). The results were contrasted with earlier studies on cardiac prediction. We intend to develop the model further in the future so that it might be applied to different feature selection strategies. Alternatively, a random forest classifier could be employed. The most important objective of this study is to enhance earlier effort by devising a novel and unique model-creation method and to render the model applicable and user-friendly in practical contexts.