Abstract

Youth is the generation that will hold the future of Indonesia. According to BPS, a quarter of Indonesia's population are youth. Thus, the government needs an overview of the current quality of youth to formulate appropriate policies for each region. This study aims to classify provinces in Indonesia based on youth data using agglomerative hierarchical and K-Means. According to the value of the internal validation and stability index, the agglomerative hierarchical, using Ward's method, with 2 clusters was chosen as the best clustering method. This method produces 2 clusters consisting of 11 and 23 provinces respectively. In general, Cluster 1 contains provinces with better youth quality, where the average youth schooling years, the percentage of youth with internet access, the percentage of youth with health insurance are higher than Cluster 2 despite having a higher unemployment rate. In contrast, Cluster 2 has a higher average score on the Youth Sickness Rate, the percentage of youth with first marriage age 16 – 18 years, and the percentage of young women who give birth to babies with LBW.


Keywords – Clustering; Hierarchical; K-Means; SDGs; Youth