K-MEDOIDS ALGORITHM CLUSTERING WITH PRINCIPAL COMPONENT ANALYSIS (PCA) (CASE STUDY: DISTRICTS/CITIES ON THE BORNEO ISLAND BASED ON POVERTY INDICATORS IN 2021)

Muhammad Yafi (Statistics Study Program, Department of Mathematics, Faculty of Mathematics and Natural Sciences, Mulawarman University, Samarinda)
Rito Goejantoro (Statistics Study Program, Department of Mathematics, Faculty of Mathematics and Natural Sciences, Mulawarman University, Samarinda)
Andrea Tri Rian Dani (Statistics Study Program, Department of Mathematics, Faculty of Mathematics and Natural Sciences, Mulawarman University, Samarinda)

Jurnal Statistika Universitas Muhammadiyah Semarang
Vol 11, No 2 (2023): Jurnal Statistika Universitas Muhammadiyah Semarang

Cluster analysis is a technique in data mining that aims to group data (object) based on the information in the data. This research is used a non-hierarchical grouping named K-Medoids algorithm to group districts/cities in Borneo island based on poverty indicators and Principal Component Analysis (PCA) method to reduce research variable. This research is also do a cluster validity test to see how many cluster there are has the best grouping result using Silhouette Coefficient (SC) method. Based on the results of the analysis there is 3 optimal Principal Component (PC) were obtained with eigen value criteria of greater than or equal to 1. Furthermore, districts/cities on Borneo island were grouped based on the PC that formed and obtained 2 optimal clusters with an SC value of 0.61. The K-Medoids algorithm obtain 2 cluster, cluster 1 consisting of 49 districts/cities and cluster 2 consisting of 7 cities.