OPTIMASI MODEL CLUSTERING DALAM PEMETAAN STUNTING DI KABUPATEN ASAHAN
DOI:
https://doi.org/10.54314/jssr.v7i4.1417Abstract
Peningkatan angka stunting di Kabupaten Asahan pada tahun 2022 mencapai 15,3%, belum memenuhi atas target nasional yang ditetapkan sebesar 14% pada tahun 2024. Untuk mengatasi masalah ini, pemerintah daerah perlu melakukan pemetaan wilayah rentan terhadap stunting menggunakan pendekatan teknologi data science yang lebih komprehensif. Namun, upaya tersebut belum dilakukan secara optimal sehingga program penanganan stunting belum efektif. Penelitian ini bertujuan untuk mengoptimalkan model clustering dalam pemetaan stunting di Kabupaten Asahan dengan menggunakan data dari berbagai puskesmas. Metode yang digunakan mencakup normalisasi data dengan Z-Score untuk mengurangi dampak outlier, penentuan jumlah klaster optimal menggunakan metode Elbow, dan inisialisasi pusat klaster menggunakan K-Means++. Hasil dari analisis menunjukkan bahwa jumlah klaster optimal adalah 3, dan evaluasi clustering menggunakan Davies-Bouldin Index (DBI) menghasilkan skor sebesar 0.589, yang menunjukkan performa model clustering yang baik. DBI yang lebih rendah mengindikasikan bahwa cluster yang terbentuk cukup kompak dan terpisah dengan baik, sehingga dapat membantu pemetaan wilayah untuk alokasi sumber daya yang lebih efisien dalam penanganan stunting di Kabupaten Asahan.
Kata kunci: Stunting; K-Means++; Elbow Method; Z-Score, Davies-Bouldin Index.
The increase in stunting rates in Asahan Regency in 2022 reached 153%, far above the national target set at 14% in 2024. To overcome this problem, local governments need to map areas vulnerable to stunting using a more comprehensive data science technology approach. However, these efforts have not been carried out optimally so that the stunting management program has not been effective. This research aims to optimize the clustering model in stunting mapping in Asahan Regency using data from various community health centers. The method used includes data normalization with Z-Score to reduce the impact of outliers, determining the optimal number of clusters using the Elbow method, and initializing cluster centers using K- Means++. The results of the analysis show that the optimal number of clusters is 3, and clustering evaluation using the Davies-Bouldin Index (DBI) produces a score of 0.589, which shows good clustering model performance. A lower DBI indicates that the clusters formed are quite compact and well separated, so they can help map areas for more efficient resource allocation in handling stunting in Asahan Regency.
 Keywords: Stunting; K-Means++; Elbow Method; Z-Score; Davies-Bouldin Index.
Â
Downloads
References
Hermawan D, Kurniasari D, Sandayanti V, Sari N, Listyaningsih E. Relationships of deworming drug consumption and animal protein intake with stunting. Parasite Epidemiol Control [Internet]. 2023;23(March):e00326. Available from: https://doi.org/10.1016/j.parepi.2023.e00326
Andriani H, Friska E, Arsyi M, Sutrisno AE, Waits A, Rahmawati ND. A multilevel analysis of the triple burden of malnutrition in Indonesia: trends and determinants from repeated cross-sectional surveys. BMC Public Health [Internet]. 2023;23(1):1–15. Available from: https://doi.org/10.1186/s12889-023-16728-y
Oyewole GJ, Thopil GA. Data clustering: application and trends [Internet]. Vol. 56, Artificial Intelligence Review. Springer Netherlands; 2023. 6439–6475 p. Available from: https://doi.org/10.1007/s10462-022-10325-y
Dalmaijer ES, Nord CL, Astle DE. Statistical power for cluster analysis. BMC Bioinformatics [Internet]. 2022;23(1):1–28. Available from: https://doi.org/10.1186/s12859-022-04675-1
Ezugwu AE, Shukla AK, Agbaje MB, Oyelade ON, José-GarcÃa A, Agushaka JO. Automatic clustering algorithms: a systematic review and bibliometric analysis of relevant literature. Neural Comput Appl [Internet]. 2021;33(11):6247–306. Available from: https://link.springer.com/article/10.1007/s00521-020-05395-4
Ikotun AM, Ezugwu AE, Abualigah L, Abuhaija B, Heming J. K-means clustering algorithms: A comprehensive review, variants analysis, and advances in the era of big data. Inf Sci (Ny) [Internet]. 2023;622:178–210. Available from: https://doi.org/10.1016/j.ins.2022.11.139
Jamshidi EJ, Yusup Y, Kayode JS, Kamaruddin MA. Detecting outliers in a univariate time series dataset using unsupervised combined statistical methods: A case study on surface water temperature. Ecol Inform [Internet]. 2022;69 (December 2021):101672. Available from: https://www.sciencedirect.com/science/article/abs/pii/S1574954122001224
