Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Early Determination of Diabetes Mellitus Disease Prediction with Decision Tree Boosting

Rektorat UDINUS 2021 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2021
Goals
Record
c43bd44b-0042-4e26-b930-62bef1aafc5e
Published
2026-10-30

Universitas Dian Nuswantoro conducted an activity on the early prediction of diabetes mellitus using decision tree boosting. The activity addressed diabetes mellitus as a disease characterized by hyperglycemia, impaired carbohydrate metabolism, and reduced performance of insulin secretion. Participants in the activity were Adhipurna, Novian; Shidik, Guruh Fajar; Fanani, A. Zainul; Muljouo; Supriyanto, Catur; Soeleman, M. Arief; and Alzami, Farrikh.

The activity focused on early prediction as an effort to prevent an increase in the number of people with diabetes mellitus. The source text states that diabetes mellitus cases amounted to 451 million in 2017 and were expected to increase to 693 million cases until 2045. To address this context, the activity implemented Adaboost to optimize the performance of a decision tree. A workflow system was designed to create a model that could analyze diabetes mellitus using boosting methods in its application.

The model was tested using six algorithms as comparisons for evaluating the analyzed results. The experiment showed an increase in the accuracy rate of DTBoost, which was reported as 98.04% higher than the previous Decision Tree accuracy level. Through this work, UDINUS addressed the stated goal of supporting early determination and prediction of diabetes mellitus through a boosting-based analytical model. The activity therefore provides an institutional example of health-related research focused on diabetes mellitus prediction and the evaluation of classification methods.

Reference https://www.scopus.com/pages/publications/85118919709?origin=resultslist