Vulnerability Mapping of Toddler Pneumonia in East Java, Indonesia, Using the K-medoids Clustering Algorithms

Abstract

In Indonesia, pneumonia is the second leading cause of death in toddler after diarrhea. East Java Province is one of the provinces with a high rate of toddler pneumonia. Pneumonia cases in East Java are the second highest in Indonesia and still a big health problem that must be solved. Mapping of toddler pneumonia susceptibility can provide an overview of the relationship between regions and the level of susceptibility. Spatial analysis can determine the pattern of disease distribution and possible causes to plan more effective community-based pneumonia efforts. This paper proposes a new approach to map the susceptibility level of toddler pneumonia in East Java using the K-medoids clustering algorithm. The attributes used as factors influencing vulnerability consisted of the number of children under five, the number of patients, the number of patients treated, healthy homes, sanitation, clean water sources, TUPM, weighed toddlers, malnutrition and poor nutrition. These criteria have a strong correlation with each other to determine the level of susceptibility of toddler pneumonia. The susceptibility level of toddler pneumonia is divided into low, medium and high levels. The mapping is visualized spatially-temporal from 2015-2019. The results of average silhouette of the K-medoids clustering 2015-2019 data generated are 0.33 better than K-means.