Perbandingan Proses Klasterisasi Data Menggunakan K-Means Clustering dan Agglomerative Hierarchical Clustering

Abstract

Large amounts of data require good processing and analysis. One of the data analysis techniques is data clustering, which is grouping data into several groups or data clusters based on the similarity of data characteristics. This study observed the clustering process and cluster results using the K-Means and Agglomerative Hierarchical Clustering (AHC) algorithms or methods. Clustering was carried out using three different amounts of data, namely 10 (A10 data), 30 (B30 data), and 60 (C60 data), with choices of two, three, and four clusters. The experimental results obtained were that the A10 data cluster was the same, but the C60 data was different. Both methods provide the same cluster results, namely in the number of cluster members and their data numbers; conversely, different cluster results are obtained if there are differences in the number of cluster members. The B30 data cluster results for three clusters are the same, while for two and four clusters they are different. The results of this study are expected to provide a better understanding of the data clustering process and can be a basis for selecting a more appropriate clustering method.