Scholarlink

The accurate prediction of diabetes diseases based on symptoms is crucial for early diagnosis and effective treatment planning. Leveraging machine learning techniques, we have developed an ensemble diabetes prediction method that considers both regular factors, such as glucose, BMI, age, and insulin, and a small number of exogenous characteristics linked to diabetes. Our study aims to analyze ten machine learning methods (Fine Tree, Medium Tree, Coarse Tree, Gaussian Naive Bayes, Fine K-NN, Medium K-NN, Coarse K-NN, Weighted K-NN, Subspace K-NN, and RUSBoosted Tree), as well as classifiers like voting and bagging classifiers, to evaluate their performance in disease prediction and their potential impact on healthcare systems. Results on the PIMA dataset demonstrate the performance of each classifier, with the Bagging Classifier achieving the highest accuracy of 82.46% and an F1-score of 82%, highlighting its potential effectiveness in diabetes prediction. Despite a slightly lower accuracy, the Voting Classifier also shows competitive performance across all evaluation metrics. This study contributes to the advancement of predictive analytics in healthcare and underscores the potential of machine learning techniques in improving patient care outcomes.

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

Ensemble Machine Learning Approaches with Voting and Bagging Classifier for Diabetes Disease Predictions

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Shital L. Sangle,
  • Hemangi Kulkarni

摘要

The accurate prediction of diabetes diseases based on symptoms is crucial for early diagnosis and effective treatment planning. Leveraging machine learning techniques, we have developed an ensemble diabetes prediction method that considers both regular factors, such as glucose, BMI, age, and insulin, and a small number of exogenous characteristics linked to diabetes. Our study aims to analyze ten machine learning methods (Fine Tree, Medium Tree, Coarse Tree, Gaussian Naive Bayes, Fine K-NN, Medium K-NN, Coarse K-NN, Weighted K-NN, Subspace K-NN, and RUSBoosted Tree), as well as classifiers like voting and bagging classifiers, to evaluate their performance in disease prediction and their potential impact on healthcare systems. Results on the PIMA dataset demonstrate the performance of each classifier, with the Bagging Classifier achieving the highest accuracy of 82.46% and an F1-score of 82%, highlighting its potential effectiveness in diabetes prediction. Despite a slightly lower accuracy, the Voting Classifier also shows competitive performance across all evaluation metrics. This study contributes to the advancement of predictive analytics in healthcare and underscores the potential of machine learning techniques in improving patient care outcomes.