EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Optimasi Algoritma SVM dengan Teknik SMOTE dan Tuning Parameter pada Klasifikasi Balita Stunting

Muhammad Al Ghorizmi Muttaqin, Gustina Alfa Trisnapradika

Building of Informatics, Technology and Science (BITS) · 08.12.2025 · 7 (3) · 1547-1556

DOI: 10.47065/bits.v7i3.8330 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Building of Informatics, Technology and Science (BITS), online erschienen am 08.12.2025. Verfasst von Muhammad Al Ghorizmi Muttaqin und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Stunting in toddlers is a chronic nutritional problem that has long-term impacts on human resource quality, including cognitive development and vulnerability to diseases. Brebes Regency is one of the priority areas for stunting management in Indonesia. This study aims to optimize the performance of the Support Vector Machine (SVM) algorithm in classifying stunting status among toddlers by addressing data imbalance using the Synthetic Minority Oversampling Technique (SMOTE) and parameter tuning. A total of 9,598 anthropometric samples collected from several community health center in Brebes were processed through stages of data cleaning, label encoding, outlier handling, standardization, and class splitting, and then divided into training data (80%) and testing data (20%). Two models were compared: the baseline SVM model and the optimized SVM model, which integrates SMOTE and parameter tuning through GridSearchCV. The results showed that the baseline model achieved an accuracy of 98.31%, but the recall for the stunting class was only 89.19%. After applying SMOTE and parameter tuning, the model’s performance improved, achieving an accuracy of 99.78% and a recall for the stunting class of 98.46%. This improvement demonstrates that the use of SMOTE and parameter tuning is highly effective in enhancing the model’s sensitivity toward the minority class. Therefore, this study shows that a comprehensive optimization approach can effectively support early detection of stunting, making it a valuable tool for more targeted health intervention planning.

Zitieren

APA

Muttaqin, M. A. G., & Trisnapradika, G. A. (2025). Optimasi Algoritma SVM dengan Teknik SMOTE dan Tuning Parameter pada Klasifikasi Balita Stunting. Building of Informatics, Technology and Science (BITS), 7(3), 1547-1556. https://doi.org/10.47065/bits.v7i3.8330

Vancouver

Muttaqin MAG, Trisnapradika GA. Optimasi Algoritma SVM dengan Teknik SMOTE dan Tuning Parameter pada Klasifikasi Balita Stunting. Building of Informatics, Technology and Science (BITS). 2025;7(3):1547-1556. doi:10.47065/bits.v7i3.8330

BibTeX
@article{Muttaqin2025,
  author = {Muttaqin, Muhammad Al Ghorizmi and Trisnapradika, Gustina Alfa},
  title = {Optimasi Algoritma SVM dengan Teknik SMOTE dan Tuning Parameter pada Klasifikasi Balita Stunting},
  journal = {Building of Informatics, Technology and Science (BITS)},
  year = {2025},
  volume = {7},
  number = {3},
  pages = {1547--1556},
  doi = {10.47065/bits.v7i3.8330}
}
Kostenlos starten →