IMPLEMENTATION OF MACHINE LEARNING ALGORITHM C4.5 IN CLASSIFICATION OF PATIENTS WITH TYPE 2 DIABETES MELLITUS - Academic study

Aug 5, 2026

Upload by Asher

The neglect of a healthy lifestyle among the Indonesian population has increased the risk of diabetes mellitus, which affects hundreds of millions of people worldwide. Early and accurate diagnosis is essential to slow disease progression. This study develops a classification model using the C4.5 machine learning algorithm to distinguish diabetic from non-diabetic individuals using diabetes-associated factors from medical records at Padang General Hospital. Evaluation reports a recall of 91%.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

INTRODUCTION

Diabetes mellitus and the need for early diagnosis

Prior work on machine learning for diabetes

RESEARCH METHODS

Diabetes (type 2 risk factors)

(Methods continue)

IMPLEMENTATION OF MACHINE LEARNING ALGORITHM C4.5 IN CLASSIFICATION OF PATIENTS WITH TYPE 2 DIABETES MELLITUS Dyah Ayu Sekar Kinasih Purwaningrum 1, Dina Agustina 2* 1,2Mathematics Department, Mathematics and Science Faculty, Padang State University Prof. Dr. Hamka Street, Air Tawar Padang, Sumatera Barat, 25132, Indonesia Corresponding author’s e-mail: * [dinagustina@fmipa.unp.ac.id](dinagustina@fmipa.unp.ac.id) ABSTRACT Article History: Received: 12th August 2023 Revised: 24th November 2023 Accepted: 25th December 2023 Keywords: C4.5 Algorithm; Classification; Diabetes Mellitus; Machine Learning. The neglect of a healthy lifestyle among the Indonesian population has led to an increased risk of diabetes mellitus, which currently affects 643 million people worldwide. Early and accurate diagnosis is crucial for preventing the progression of the disease. This study utilized the C4.5 machine learning algorithm to develop a model to classify individuals as diabetic or nondiabetic based on diabetes-associated factors. The data used in this research consisted of medical records from patients with and without diabetes at Padang General Hospital. The model's performance evaluation resulted in a recall value of 91%. By promoting a healthy lifestyle and raising awareness about the importance of regular check-ups, the burden of diabetes can be reduced, and the overall health of the population can be improved. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-ShareAlike 4.0 International License. How to cite this article: D. A. S. K. Purwaningrum and D. Agustina.,“IMPLEMENTATION OF MACHINE LEARNING ALGORITHM C4.5 IN CLASSIFICATION OF PATIENTS WITH TYPE 2 DIABETES MELLITUS,” BAREKENG: J. Math. & App., vol. 18, iss. 1, pp. 0193-0204, March, 2024. Copyright © 2024 Author(s) Journal homepage: [https://ojs3.unpatti.ac.id/index.php/barekeng/](https://ojs3.unpatti.ac.id/index.php/barekeng/) Journal e-mail: [barekeng.math@yahoo.com](barekeng.math@yahoo.com); [barekeng.journal@mail.unpatti.ac.id](barekeng.journal@mail.unpatti.ac.id) Research Article ∙ Open Access 1. INTRODUCTION The adoption of a healthy lifestyle can significantly reduce the risk of various diseases [1] . However, many people in Indonesia still tend to overlook this aspect; this fact is supported by a survey conducted by AIA Group across 15 Asia-Pacific countries, which ranked Indonesia 11th in terms of the adoption of a healthy lifestyle [2] . Consequently, the Indonesian population remains vulnerable to various diseases, one of which is diabetes mellitus. Diabetes mellitus, commonly referred to as diabetes, is a disease primarily caused by hyperglycemia, which is the accumulation of glucose in the bloodstream. Nevertheless, it is believed that the cause of type 2 diabetes is also influenced by factors, such as excessive body mass index, aging, and family history [3] . If the accumulation of glucose is due to the immune system attacking the pancreas’s insulin-producing cells, it is classified as type 1 diabetes. On the other hand, when insufficient insulin is produced , or the body’s cells do not effectively utilize the insulin hormone, it is categorized as type 2 diabetes [4] . The number of cases for both types of diabetes continues to rise, but type 2 diabetes constitutes a larger ratio, accounting for 90% of all diabetes cases [3] . As of 2021 , approximately 643 million individuals worldwide were affected by diabetes. Indonesia ranked fifth among countries with the highest number of diabetes patients, reaching 19.5 million cases, and it is predicted to increase further to 28.6 million by 2045 [3] . The alarming statistics necessitate various efforts to prevent the escalating number of diabetes patients, and one of these approaches involves utilizing medical record data [5] . In the realm of healthcare, a wealth of medical record data exists that could become i

FAQs

What problem does the study address?

The study targets early and accurate diagnosis of type 2 diabetes by distinguishing diabetic and non-diabetic individuals using diabetes-related risk factors.

How is the C4.5 algorithm used in this research?

C4.5 is used to build a classification model trained on medical records to predict whether a person is diabetic or non-diabetic.

What performance result does the study report?

The model evaluation yields a recall value of 91%, indicating strong sensitivity in identifying diabetic cases.

Aug 5, 2026

Upload by Asher