PERBANDINGAN METODE NEAREST NEIGHBOR, WARD DAN K-MEANS DALAM MENENTUKAN CLUSTER DATA KINERJA KANTOR UNIT BANK ABC
Abstract
This study aims to determine the steps for clustering analysis using Nearest Neighbor, Ward and K-Means method, and compare the results of these three methods for clustering performance data of Unit Office of ABC Bank.
Cluster analysis was performed using IBM SPSS Version 23. Results of the three methods were compared using standard deviation ratio in the group (Sw) and the standard deviation between groups (Sb). In this study, based on the ratio of standard deviation in the cluster and the standard deviation between clusters show that the Ward’s method has a better performance when used in clustering unit’s performance data of ABC Bank, this is indicated by the value of the ratio (Sw / Sb) is the smallest among the three methods that is equal to 0.353302.
Thus, the results of the cluster analysis that will be used as a reference in determining the performance data cluster of unit office of ABC Bank is the result of cluster analysis using Ward method.
Downloads
References
Kinerja Metode Ward dan K-Means dalam Menentukan Cluster Data Mahasiswa Pemohon Beasiswa (Studi Kasus : STMIK Pringsewu). Tesis. Program Pascasarjana Magister Teknik Informatika IBI Darmajaya. Bandar Lampung.
[2] Kusrini, dan Luthfi, E.T. 2009.
Algoritma Data Mining. Yogyakarta: Penerbit Andi.
[3] Han, Jiawei, Kamber, M., and Pei,
Jian. 2011. Data Mining: Concepts and Techniques, 3rd edition. Morgan Kaufmann Publishers.
[4] Hermawati, F.A. 2013. Data Mining.
Yogyakarta: Penerbit Andi.
[5] Simamora, Bilson. 2004. Riset
Pemasaran. Jakarta: Gramedia
Utama.
[6] Prasetyo, Eko. 2014. Data Mining :
Mengolah Data Menjadi Informasi Menggunakan Matlab. Yogyakarta: Penerbit Andi.
[7] Arai, Kohei and Barakbah, A. R. 2007.
Hierarchical K-means : An Algorithm for Centroids Initialization for K-Means. Reports of the Faculty of Science and Engineering, Saga University, Vol.36, No.1, 25-31.
