Comparative Analysis of Hierarchical Clustering and K-Medoids for Clustering Cases of Childhood Respiratory Diseases in Lamongan Regency
DOI:
https://doi.org/10.33005/jasid.v2i01.37Keywords:
Pediatric Respiratory Diseases, Hierarchical Clustering, Ward Linkage, K-MedoidsAbstract
Abstract— Respiratory diseases affecting children remain a significant health issue in Indonesia, including in Lamongan Regency. The region faces challenges related to pediatric respiratory illnesses, particularly Childhood Tuberculosis, Pneumonia in toddlers, and Cough in toddlers, which impact children's quality of life and development. Therefore, understanding the spatial distribution and correlation patterns among these diseases is essential to support more targeted health intervention planning. This study analyzes the distribution patterns of pediatric respiratory diseases in Lamongan Regency and clusters regions based on similarities in the number of cases using an unsupervised learning approach. The method employed is Hierarchical Clustering with four distance calculation techniques: single, complete, average, and ward linkage and K-Medoids with two distance calculation techniques: euclidean and manhattan distance. The data, sourced from the Lamongan District Health Office, include four numerical variables related to respiratory diseases, aggregated by sub-districts. Data normalization was carried out using standardization, and cluster quality was evaluated using three internal metrics: Silhouette Score, Davies-Bouldin Index (DBI), and Calinski-Harabasz Index (CHI). The analysis results indicate that the optimal number of clusters is three. Among all methods tested, the Hierarchical Clustering with ward linkage method yielded the best performance, with a Silhouette Score of 0.5447, a DBI of 0.5884, and a CHI of 20.3018. These results demonstrate that the ward linkage method is the most effective in clustering regions based on the characteristics of pediatric respiratory disease cases and can be used for mapping priority health intervention areas in Lamongan Regency.
References
Kementerian Kesehatan Republik Indonesia, Profil Kesehatan Indonesia Tahun 2022. Jakarta: Kemenkes RI, 2023. [Online]. Tersedia: https://www.kemkes.go.id/resources/download/pusdatin/profil-kesehatan-indonesia/
D. Restiana, A. Ramadhan, dan S. P. Mahendra, “Implementasi Metode K-Means dan K-Medoids untuk Klasterisasi Penyakit Berdasarkan Gejala Pasien,” Jurnal Teknologi dan Sistem Komputer, vol. 9, no. 1, pp. 40–45, 2021, doi: 10.14710/jtsiskom.9.1.2021.40-45.
P.-N. Tan, M. Steinbach, A. Karpatne, dan V. Kumar, Introduction to Data Mining, 2nd ed. Boston, MA: Pearson, 2019.
L. Kaufman dan P. J. Rousseeuw, Finding Groups in Data: An Introduction to Cluster Analysis. Hoboken, NJ: John Wiley & Sons, 2009.
J. Han, M. Kamber, dan J. Pei, Data Mining: Concepts and Techniques, 3rd ed. Waltham, MA: Morgan Kaufmann, 2012.
T. Hastie, R. Tibshirani, dan J. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd ed. New York: Springer, 2009.
D. L. Davies dan D. W. Bouldin, “A cluster separation measure,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-1, no. 2, pp. 224–227, Apr. 1979, doi: 10.1109/TPAMI.1979.4766909.
T. Calinski dan J. Harabasz, “A dendrite method for cluster analysis,” Communications in Statistics, vol. 3, no. 1, pp. 1–27, 1974, doi: 10.1080/03610927408827101.
S. Tuhpatussania, S. Erniwati, dan Z. Mutaqin, “Perbandingan metode agglomerative hierarchical clustering dan metode K‑Medoids dalam pengelompokan data titik panas kebakaran hutan di Indonesia,” Journal Computer and Technology, vol. 2, no. 1, pp. 21–38, Jul. 2024, doi: 10.69916/comtechno.v2i1.146.
G. R. Suraya dan A. W. Wijayanto, “Comparison of Hierarchical Clustering, K Means, K Medoids, and Fuzzy C Means methods in grouping provinces in Indonesia according to the special index for handling stunting,” Indonesian Journal of Statistics and Its Applications, vol. 6, no. 2, pp. 180–201, Aug. 2022, doi: 10.29244/ijsa.v6i2p180-201.
