IMPLEMENTATION OF CLUSTERING USING K-MEANS METHOD TO DETERMINE NUTRITIONAL STATUS

Abstract

Cluster analysis aims to classify data objects into two categories: objects that are similar in characteristics in one cluster and objects that are different in characteristics with the other objects of another cluster. K-Means is a method included in the distance-based clustering algorithm that starts by determining the number of desired clusters. Malnutrition is one of the biggest concerns in Indonesia. According to Riskesdas 2018 data, as many as 17.7% infants under 60-month-old are still having problems with nutrition intake while 3.9% are having malnutrition. This might result in higher death rate. This research was conducted to classify the nutritional status of infants under 60-month-old conducted by the C-Means Clustering method. This research is non-reactive, using secondary data in Ponkesdes Mayangrejo, Bojonegoro without direct interaction with the subject. This study concluded that the grouping of nutritional status is possible by using K-Means with 4 clusters formed which are 23 malnourished toddlers, 17 undernourished toddlers, 7 nourished toddlers, and 10 over-nourished toddlers.