PENGELOMPOKAN KABUPATEN/KOTA DI PROVINSI JAWA TENGAH BERDASARKAN INDIKATOR KESEHATAN BAYI DAN BALITA MENGGUNAKAN ALGORITMA FUZZY C-MEANS DAN K-MEDOIDS

Abstract

The infant and toddler mortality rate is an indicator of health in Indonesia. If the health level of infants and toddlers in an area is low, the health status in that area is low. Clustering of districts based on infant and toddler health indicators can be used as a guideline for the government in allocating funds and determining health services program. Clustering is a statistical data processing technique that is useful for grouping an area into a klaster based on certain characteristics. This clustering of infants and toddlers uses Fuzzy C- Means and K-Medoids algorithms. The optimal cluster of the two methods is selected using silhouette validation, while for selection of the best method for profiling using standard deviation ratio values. The optimal number of clusters in the Fuzzy C-Means and K-Medoids algorithms based on silhouette validation for each method is 2 clusters. Based on the value of the standard deviation ratio of each method, the value of Fuzzy C-Means (1.1062) is smaller than K-Medoids (1,1771), so the 2-klaster of Fuzzy C-Means method is used for the profiling step. Cluster 1 is a group of regions with infant and toddler mortality rates, malnutrition and low birth weight babies, so there must be an increase in the complete basic immunization program and health services. Cluster 2 is a regional group with very good health services, so it is better to maintain the quality of these health services. 1.