Penerapan Metode Agglomerative Clustering Untuk Segmentasi Data Dalam Lingkungan Big Data

Paskal Arienda Epindonta Ginting (Ilmu Komputer, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Negeri Medan)
Risky Immanuel Situmorang (Ilmu Komputer, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Negeri Medan)
Muhammad Raihansyah Lubis (Ilmu Komputer, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Negeri Medan)
Raja Ansel Hartama Sihombing (Ilmu Komputer, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Negeri Medan)
Arnita Piliang (Ilmu Komputer, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Negeri Medan)

Jurnal Sistem Informasi dan Informatika
Vol 4 No 1 (2026): Januari 2026

The exponential growth of data in the digital era has increased the need for analytical methods capable of handling Big Data characteristics. This study examines the application of Agglomerative Hierarchical Clustering (AHC) for data segmentation using two datasets: (1) an Iris dataset of 24 samples with 8 morphological attributes, and (2) an e-commerce transaction dataset of 10 customer records. Ward linkage was selected based on literature evidence of its superiority. Results on the Iris dataset yielded 3 optimal clusters with a Silhouette Score of 0.4196 and an Adjusted Rand Index of 0.3635, achieving 70.83% classification accuracy. In the e-commerce dataset, three customer segments were formed: premium, middle-tier, and passive customers. These findings confirm AHC as an effective multidimensional data segmentation method.