Scholarlink

Diabetes is a chronic illness, which causes a severe effect on the health of human beings. In this disease, the sugar level in blood remains higher than the normal. It happens when either the pancreas is not able to make adequate insulin or the cells in the body are not efficiently utilizing it. This disease may damage different body organs such as eyes, kidneys and may even lead to heart diseases. In this research work, we used the Attribute Subset selection for feature extraction and dimensionality reduction, K-means algorithm for removal of impure instances as outliers. Logistic regression approach is used for classification, which classifies input data and predict the diabetes class. The Attribute Subset Selection is carried out with the assistance of Weka tool kit and then rest of the proposed approach is implemented in Python and outcomes are evaluated in terms of recall, accuracy and precision parameters.

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

A Novel Approach for Diabetic Prediction Using Attribute Subset Selection, K-Means and Logistic Regression

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Kumar Gaurav Upadhyay,
  • Mala Kalra,
  • Rakesh Kumar,
  • Prashant Kumar

摘要

Diabetes is a chronic illness, which causes a severe effect on the health of human beings. In this disease, the sugar level in blood remains higher than the normal. It happens when either the pancreas is not able to make adequate insulin or the cells in the body are not efficiently utilizing it. This disease may damage different body organs such as eyes, kidneys and may even lead to heart diseases. In this research work, we used the Attribute Subset selection for feature extraction and dimensionality reduction, K-means algorithm for removal of impure instances as outliers. Logistic regression approach is used for classification, which classifies input data and predict the diabetes class. The Attribute Subset Selection is carried out with the assistance of Weka tool kit and then rest of the proposed approach is implemented in Python and outcomes are evaluated in terms of recall, accuracy and precision parameters.