Optimasi Centroid Awal Algoritma K-Medoids Menggunakan Particle Swarm Optimization Untuk Segmentasi Customer

Authors

  • Danang Bagus Wijaya Dian Nuswantoro University
  • Edi Noersasongko Dian Nuswantoro University
  • Purwanto Purwanto Dian Nuswantoro University

DOI:

https://doi.org/10.62411/tc.v23i1.9516

Keywords:

Customer segmentation, k-medoids, Particle Swarm Optimization, Davies-Bouldin Index

Abstract

Customer segmentation is an important strategy in a company, it affects good customer relationships which will result in increased profits. Grouping customers in data mining can use several algorithms, but K-Medoids is the right choice because it can reduce noise and outlier sensitivity. However, the selection of cluster centers is still random and has an effect on the results of clustering, so it is necessary to improve the k-medoids algorithm so that the resulting cluster value can be optimal. Particle Swarm Optimization is an optimization algorithm that is often used and has been proven to improve the results of a clustering. In this case, optimization using Particle Swarm Optimization (PSO) in the selection of the initial cluster center needs to be applied to the k-medoids algorithm so that the results of the cluster can be optimal. The results of the study showed the Davies-Bouldin Index (DBI) value for K-Medoids K 2 = 0.379, K 3 = 0.283, and K 4 = 0.593, while the DBI value PSO + K-Medoids K 2 = 0.088, K 3 = 0.226, and K4 = 0.363. The DBI value shows that PSO optimization on K-Medoids to determine the initial centroid is proven to improve the results of clustering than standard K-Medoids.

References

B. E. Adiana, I. Soesanti and A. E. Permanasari, "Analisis Segmentasi Pelanggan Menggunakan Kombinasi Rfm Model dan Teknik Clustering," JUTEI (Jurnal Terapan Teknologi Informasi), pp. 23-32, 2018.

R. W. B. S. Berahmana, F. A. Mohammed and K. Chairuang, "Customer Segmentation Based on RFM Model Using K-Means, K-Medoids, and DBSCAN Methods," Lontar Komput. J. Ilm. Teknol. Inf., vol. 11, no. 1, p. 32–43, 2020.

P. Kotler, B. Molan, B. Sarwiji and K. Lane, Marketing Management, Thirteen Edition, Jakarta: Erlangga, 2009.

A. A. D. Sulistyawati and M. Sadikin, "Penerapan Algoritma K-Medoids Untuk Menentukan Segmentasi Pelanggan," SISTEMASI: Jurnal Sistem Informasi, vol. 10, no. 3, pp. 516-526, 2021.

D. Yu, L. Guojun, M. Guo and X. Liu, "An improved K-Medoids based on step increasing and optimizing medoids," Expert System with Applications, vol. 92, pp. 464-473, 2018.

S. K. Majhi and S. Biswal, "Optimal cluster analysis using hybrid K-Means and Ant Lion Optimizer," Karbala International Journal of Modern Science, vol. 4, no. 4, pp. 347-360, 2018.

C. Mageshkuma, S. Karthik and P. Arunachalam, "Hybrid metaheuristic algorithm for improving the efficiency of data clustering," Cluster Computing, 2018.

J. Kennedy and R. Eberhart, "Particle swarm optimization," In Proceedings of the 1995 IEEE International Conference on Neural Networks. IEEE Service Center, Piscataway, 1995.

C. P. Lekshmy and N. Abdul, "An Improved Clustering Algorithm based on K-Means and Harmony Search Optimization," IEEE, 2011.

D. T. Larose, Discovering Knowledge in Data: An Introduction to Data Mining, Wiley, 2005.

D. F. Pramesti, M. T. Furqon and C. Dewi, "Implementasi Metode K-Medoids Clustering Untuk Pengelompokan Data Potensi Kebakaran Hutan/Lahan Berdasarkan Persebaran Titik Panas (Hotspot)," Pengembangan Teknologi Informasi dan Ilmu Komputer, vol. 1, no. 9, pp. 723-732, 2017.

S. S. Rao, Engineering Optimization, Theory and Practice. fourth edition, New York: John Wiley & Sons, 2009.

D. L. Davies and D. W. Bouldin, "A Cluster Separation Measure," EEE Trans. Pattern Anal. Mach. Intell.,, vol. 1, no. 2A, 1979.

Downloads

Published

2024-06-18