The Implementation of the K-Means Clustering Algorithm Based on the Severity Level of Diabetes in Patients Using a Website Platform

Abstract

Diabetes mellitus is a chronic non-communicable disease with a steadily increasing prevalence worldwide, posing a significant public health challenge due to its potential for severe complications if not managed properly. In many healthcare facilities, including RSUD Pariaman, there is still no structured system to classify patients according to the severity of their condition, which hampers timely intervention and optimal resource allocation. This study aims to develop and implement a web-based system for clustering the severity levels of type 2 diabetes mellitus using the K-Means Clustering algorithm as a decision support tool for medical staff. A quantitative system development research design was applied, utilizing secondary medical records from January 2023 to December 2024, with five clinical variables: Hemoglobin A1c (HbA1c), Fasting Blood Glucose (GDP), Systolic Blood Pressure (TDS), Diastolic Blood Pressure (TDD), and Body Mass Index (BMI). The system was built using the CodeIgniter PHP framework, MySQL database, and Bootstrap-based interface, following the Knowledge Discovery in Database (KDD) process for data preprocessing. K-Means clustering was configured into three categories (mild, moderate, and severe). Validation using RapidMiner confirmed that the clustering results from the web-based system were consistent with the benchmark model, ensuring the correctness of the algorithm’s implementation. The developed system enables real-time data processing, displays results in both tabular and graphical forms, and provides an intuitive interface for medical personnel, thus supporting clinical decision-making and improving healthcare service quality.