Comparing K-Means and DBSCAN Algorithms for Clustering Poverty Levels in Papua Islands

Poverty has become one of the main challenges in Indonesia, especially in the eastern regions. According to data from the Central Bureau of Statistics (BPS), Papua Island currently ranks as the poorest province. As of 2018, the poverty rate in Papua was 27.6%, and as of March 2023, it was 26.03%. The government has made various efforts to address the issue of poverty in Papua, one of which is providing financial assistance in the form of BLT (Direct Cash Assistance). However, to achieve sustainable development, more effective and targeted strategies are needed. The government's efforts to address poverty in Papua must be supported by accurate and efficient data analysis. This study involves clustering data on impoverished populations from 42 districts/cities in Papua and West Papua, obtained from the Central Bureau of Statistics (BPS) website in 2023. The study aims to compare the clustering results using the K-Means and DBSCAN algorithms. The performance of these algorithms is determined by the highest Silhouette Index (SI) value. After the data underwent a standardization process, the clustering results using the K-Means algorithm achieved the best SI value of 0.6359 with K=4. Meanwhile, the clustering results using the DBSCAN algorithm achieved the best SI value of 0.3678 with eps 1.8 and MinPts 8. The SI value of the K-Means algorithm is higher than that of the DBSCAN algorithm, indicating that clustering using the K-Means algorithm is superior to DBSCAN.