PERBANDINGAN SELF ORGANIZING MAP DAN K-AFFINITY PROPAGATION DALAM MENGELOMPOKAN KASUS STUNTING

Abstrak

Stunting is a physical failure in toddlers due to chronic malnutrition so that children are too short for their age. One of the provinces with the highest cases of stunting in Indonesia is East Nusa Tenggara (NTT) Province. Based on the results of the Indonesian Nutritional Status Survey (SSGI) on the prevalence of stunting in 2022, NTT province is in first place with a figure of 35.5%. One effort that can be made to minimize stunting rates and make human resources efficient is by focusing interventions on areas that have a significant impact on the high prevalence of stunting. This can be done by using clustering analysis. Clustering is the process of grouping data into several groups so that data in one group has a maximum level of similarity and data between groups has a minimum level of similarity. This study implemented clustering techniques using SOM and K-AP, because several variables in the data in this study contain outliers and both methods are not sensitive to outliers in the data so that they can produce optimal clusters even in data containing outliers. Then a comparison was made between the SOM and K-AP methods. The results of data analysis using the SOM and K-AP methods showed that the optimal method for grouping stunting cases based on districts/cities in NTT province in 2022 was the K-AP clustering method. This is based on the smallest variance value, namely 0.009469122.