Enhanced Diabetes Risk Classification Using Broad Learning System and SMOTE-ENN Balancing
JOIV : International Journal on Informatics Visualization · 31.07.2026 · 10 (4) · 1568
DOI: 10.62527/joiv.10.4.4808 ↗Kurz gefasst
Fachartikel in der Zeitschrift JOIV : International Journal on Informatics Visualization, online erschienen am 31.07.2026. Verfasst von Muhammad Ijaz Shafarin und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Themen: Diabetes mellitus
Abstract (Originalsprache)
This study addresses the need for a learning method to support more accurate diagnosis and classification of diabetes, a chronic metabolic disorder that can lead to complications. This study investigates machine learning models, specifically the Broad Learning System model, to assess the effectiveness of model development and classification results in identifying imbalanced patterns in medical data. In this study, the Broad Learning System (BLS) method plays the main role, and a BLS variant, Tree-BLS, is used as a comparison method to leverage feature diversity. The BLS model has a component in expanding the network horizontally, while Tree-BLS has a tree-shaped structure component to enrich feature representation. A significant finding in this study is the results shown from the BLS model, which shows an accurate rate of 92.97% in the average model evaluation using k-fold cross-validation. In addition, the precision is 92.79%, the recall is 95.29%, and the F1 score is 93.86%. This BLS machine learning model shows potential to advance artificial intelligence technology that could eventually become a diabetes prediction service in an ideal health system. The results emphasize the importance of stable learning methods and datasets to improve diagnostics and support continuous progress in the healthcare system toward more accurate, reliable, and effective diabetes prediction.
Zitieren
APA
Shafarin, M. I., Pertiwi, K. M. D., & Saputra, D. C. E. (2026). Enhanced Diabetes Risk Classification Using Broad Learning System and SMOTE-ENN Balancing. JOIV : International Journal on Informatics Visualization, 10(4), 1568. https://doi.org/10.62527/joiv.10.4.4808
Vancouver
Shafarin MI, Pertiwi KMD, Saputra DCE. Enhanced Diabetes Risk Classification Using Broad Learning System and SMOTE-ENN Balancing. JOIV : International Journal on Informatics Visualization. 2026;10(4):1568. doi:10.62527/joiv.10.4.4808
BibTeX
@article{Shafarin2026,
author = {Shafarin, Muhammad Ijaz and Pertiwi, Kharisma Monika Dian and Saputra, Dimas Chaerul Ekty},
title = {Enhanced Diabetes Risk Classification Using Broad Learning System and SMOTE-ENN Balancing},
journal = {JOIV : International Journal on Informatics Visualization},
year = {2026},
volume = {10},
number = {4},
pages = {1568},
doi = {10.62527/joiv.10.4.4808}
}
EUVIMED