본문 바로가기
  • Home

Design and Analysis of Dual Weighted KNN Algorithm_Based Health Status Prediction System

  • Journal of Software Assessment and Valuation
  • Abbr : JSAV
  • 2025, 21(2), pp.123~129
  • Publisher : Korea Software Assessment and Valuation Society
  • Research Area : Engineering > Computer Science
  • Received : June 6, 2025
  • Accepted : June 20, 2025
  • Published : June 30, 2025

Jeong Yon Shim 1

1강남대학교

Accredited

ABSTRACT

Early prediction of chronic diseases have become increasingly critical in modern healthcare. To support this need, automated health prediction systems using biosignals such as heart rate, blood pressure, and blood glucose levels have been actively studied. Among various classification methods, KNN is widely used due to its simplicity and effectiveness in certain scenarios. However, conventional KNN treats all features as equally important, which may conflict with actual clinical priorities. In this study, we propose a dual-weighted KNN classifier that integrates both distance-based weights and feature-based weights. In particular, the model assigns higher importance to blood glucose levels. We also develop a system that enables real-time health status prediction for user input and evaluate the model across various values of k. Experimental results demonstrate that the proposed approach improves prediction accuracy compared to traditional KNN, validating its effectiveness for health monitoring applications

Citation status

* References for papers published after 2023 are currently being built.