EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Data-Driven K-Means Clustering Analysis for Stunting Risk Profiling of Pregnant Women

Desvita Dian Nazella, Heru Pramono Hadi, Farrikh Al Zami, Ayu Ashari, Yupie Kusumawati, Suharnawi Suharnawi, Rama Aria Megantara, Muhammad Naufal

Building of Informatics, Technology and Science (BITS) · 08.12.2025 · 7 (3) · 1628-1636

DOI: 10.47065/bits.v7i3.8415 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Building of Informatics, Technology and Science (BITS), online erschienen am 08.12.2025. Verfasst von Desvita Dian Nazella und 7 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Stunting in children is influenced by maternal health conditions during pregnancy. This study aims to classify pregnant women to prevent stunting based on clinical, demographic, and environmental factors using the K-Means Clustering algorithm. A total of 229 data from the Primadona application (Disdalduk KB Kota Semarang) were analyzed using 14 normalized variables. The optimal number of clusters was determined using the Elbow Method and validated using the Silhouette Score, Davies-Bouldin Index, and Calinski-Harabasz Index. The Kruskal-Wallis test was performed to verify differences between clusters. This study resulted in seven clusters with different profiles, with a Silhouette Score of 0.134, Davies-Bouldin Index of 1.509, and Calinski-Harabasz Index of 29.54. These values ​​indicate that the cluster structure is formed and reflects the variation in risk for pregnant women, although there is overlap due to differences in characteristics between individuals. The clustering successfully differentiated pregnant women with low to high risk, influenced by health and environmental factors. This study proves the effectiveness of K-Means in identifying stunting risk patterns in pregnant women and supports more targeted interventions, such as nutritional counseling, disease risk monitoring, education on cigarette smoke exposure, and referrals. Limitations of this study include the unbalanced distribution of data between and the use of cross-sectional data. Future research is recommended to improve pre-processing and compare other clustering methods such as K-Medoids or DBSCAN for more precise stunting risk analysis.

Zitieren

APA

Nazella, D. D., Hadi, H. P., Zami, F. A., Ashari, A., Kusumawati, Y., Suharnawi, S., Megantara, R. A., & Naufal, M. (2025). Data-Driven K-Means Clustering Analysis for Stunting Risk Profiling of Pregnant Women. Building of Informatics, Technology and Science (BITS), 7(3), 1628-1636. https://doi.org/10.47065/bits.v7i3.8415

Vancouver

Nazella DD, Hadi HP, Zami FA, Ashari A, Kusumawati Y, Suharnawi S, et al. Data-Driven K-Means Clustering Analysis for Stunting Risk Profiling of Pregnant Women. Building of Informatics, Technology and Science (BITS). 2025;7(3):1628-1636. doi:10.47065/bits.v7i3.8415

BibTeX
@article{Nazella2025,
  author = {Nazella, Desvita Dian and Hadi, Heru Pramono and Zami, Farrikh Al and Ashari, Ayu and Kusumawati, Yupie and Suharnawi, Suharnawi and Megantara, Rama Aria and Naufal, Muhammad},
  title = {Data-Driven K-Means Clustering Analysis for Stunting Risk Profiling of Pregnant Women},
  journal = {Building of Informatics, Technology and Science (BITS)},
  year = {2025},
  volume = {7},
  number = {3},
  pages = {1628--1636},
  doi = {10.47065/bits.v7i3.8415}
}
Kostenlos starten →