Implementation of K-Means Clustering and Agglomerative Hierarchical Clustering Methods to Determine the Nutritional Status of Toddler

Abstract

Abstract

Problems related to fulfilling toddler nutrition are still homework in Indonesia. The results of the survey data on the  nutritional  status  of  children  under  five  in Indonesia (SSGBI)  for  2021  tackling  the prevalence  of  stunting  in Indonesia reached 24.4%, wasting reached 7.1%, and wasting reached 17.0%. The number of stunting toddlers in Indonesia still exceeds the WHO threshold, which is 20%. Therefore, to reduce the level of malnutrition, it is necessary to record and classify the nutritional status of children under five. This research aim to grouping the data of toddlers based on the age by month and weight by kg using KnA algorithm. By using the silhouette score, 2 clusters has the closest value to 1 compared to the number of other clusters, where cluster 0 has a total of 69 toddlers, while the number of toddlers in cluster 1 is 112 toddlers. This result can be used for posyandu to analyze toddler segmentation.