Scholarlink

In the present scenario, diabetes problem is a chronic disease, which extensively contributes to global ill health. The possibility of early diagnosis would allow for timely interference and administration of problems arising from diabetes. Presented paper explores the use of various types of ML methods for diagnosing of diabetes, employing various health indicators, glucose, blood pressure, and BMI, among others. The performance of four algorithms—logistic regression, random forest, decision trees, and neural network—is compared to analyze the results using the Pima Indians Diabetes dataset. The aim of this presented work is to find out the most effective methods for diabetes prediction coupled with a major emphasis on the importance of the feature selection technique in achieving better accuracy of the model.

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

Enhancing Diabetes Prediction Accuracy with Feature Engineering Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Rakesh Kumar Saxena,
  • Sunil Kushwaha,
  • Satya Prakash Awasthi,
  • Vinita Kushwaha

摘要

In the present scenario, diabetes problem is a chronic disease, which extensively contributes to global ill health. The possibility of early diagnosis would allow for timely interference and administration of problems arising from diabetes. Presented paper explores the use of various types of ML methods for diagnosing of diabetes, employing various health indicators, glucose, blood pressure, and BMI, among others. The performance of four algorithms—logistic regression, random forest, decision trees, and neural network—is compared to analyze the results using the Pima Indians Diabetes dataset. The aim of this presented work is to find out the most effective methods for diabetes prediction coupled with a major emphasis on the importance of the feature selection technique in achieving better accuracy of the model.