Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

A Weighted Majority Voting Ensemble Model for Diabetes Prediction Using PSO

Rektorat UDINUS 2023 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2023
Goals
Record
ab91c3c8-b809-43c3-ae65-b69524b89c5c
Published
2026-10-30

Universitas Dian Nuswantoro conducted research to develop a model for predicting diabetes mellitus using the Pima Indian Diabetes Dataset (PIDD). The activity involved researchers from UDINUS, including MULJONO, ADHIPURNA, and NOVIAN. The research addressed the importance of early detection of diabetes in the context of the continuing rise in the number of cases. It also considered limitations in existing approaches to diagnosis, particularly their effectiveness and efficiency in supporting clinical decisions.

The activity used an intelligent system to assist with data analysis and clinical decision support. An ensemble method based on majority voting was examined because ensemble methods have been shown to be effective in increasing accuracy. The research addressed the issue of determining the optimal weight for the ensemble voting process. To improve prediction accuracy, Particle Swarm Optimization, identified in the source as PSO, was employed in selecting the weights for ensemble majority voting.

The test results reported that predictions from ensemble majority voting using PSO performed well. The reported accuracy was 0.9322, precision was 0.9412, recall was 0.8421, and the F1-score was 0.8889. The source also states that PSO accuracy improved by 4% and 7%, respectively. These findings demonstrate, according to the source, that applying PSO to the ensemble model can improve accuracy. The activity relates to the health focus in RAW_GOALS through research on diabetes prediction, early detection, data analysis, and clinical decision support. It presents UDINUS research activity focused on developing and testing an intelligent model for diabetes mellitus prediction.

Reference https://www.scopus.com/pages/publications/85185289861?origin=resultslist