This study aims to compare the effectiveness of two clustering algorithms—K-Means
and Hierarchical Clustering—in grouping Indonesian provinces based on key labor
indicators: the Open Unemployment Rate (TPT) and the Labor Force Participation Rate
(TPAK). The study uses secondary data from Indonesia’s Central Bureau of Statistics (BPS)
covering the years 2019–2023. The analysis follows the CRISP-DM framework, which
includes business understanding, data understanding, data preparation, modeling,
evaluation, and deployment phases. The optimal number of clusters is determined using the
Elbow Method for K-Means and dendrogram analysis for Hierarchical Clustering.
Clustering performance is evaluated using the Davies-Bouldin Index (DBI). The results show
that Hierarchical Clustering yields a more optimal clustering result with a DBI score of
0.8064, compared to 0.8872 for K-Means. Both methods successfully divide Indonesian
provinces into two main clusters with distinct labor characteristics. This research is expected
to support the development of more targeted employment policies and serve as a reference
for future data mining studies in the socio-economic field.