Metode Pengclusteran Berbasis Densitas Menggunakan Algoritma DBSCAN

Nur Arsih, Nusar Hajarisman, Sutawanir Darwis

Abstract


Abstract. Cluster analysis (clustering) is one of the statistical technique used to classify a range of data with high similarity in comparison to one another, but they are different from the objects in other groups. Clustering generally classified into hierarchical and non-hierarchical algorithm later evolved into many of them DBSCAN. DBSCAN is one algorithm which classifies the object based on the density of the input parameters Eps and MinPts. In this paper DBSCAN method will be compare with the k-means. The data used is secondary data of customers that having a credit facility. The results show that with Eps = 0.0128 and MinPts = 5 debtor data is divided into two clusters, that is good credit cluster and bad credit cluster with a run time faster than classical k-means algorithm.

 

Abstrak. Analisis cluster (clustering) merupakan salah satu teknik statistika yang digunakan untuk mengelompokkan suatu gugus data dengan kemiripan yang tinggi dibandingkan satu sama lain, tetapi mereka berbeda dengan objek di lain kelompok. Umumnya clustering diklasifikasi menjadi hierarki dan non hierarki yang kemudian berkembang menjadi banyak algoritma diantaranya DBSCAN. DBSCAN merupakan salah satu algoritma yang mengelompokkan objek berdasarkan densitas dengan parameter input Eps dan MinPts. Dalam skripsi ini metode DBSCAN akan dibandingkan dengan k-means. Adapun data yang digunakan adalah data sekunder nasabah yang memiliki fasilitas kredit. Hasilnya menunjukkan bahwa dengan Eps=0,0128 dan MinPts=5 data debitur terbagi menjadi dua cluster yaitu cluster kredit baik dan buruk dengan run time lebih cepat dibanding algoritma klasik k-means


Keywords


Cluster Analysis, DBSCAN, Density, Eps, K-means, Min Pts, Noise.

References


Anindya Santika Devi, N. M., Gede Darma Putra, I. K., & Sukarsa, I. (2015). Implementasi Metode Clustering DBSCAN pada Proses Pengambilan Keputusan. Lontar Komputer, 6(3).

B. Santosa. (2007). Data Mining. Teknik Pemanfaatan Data untuk Keperluan Bisnis, First Edition ed. Yogyakarta: Graha Ilmu.

Ester, M., Kriegel, H.P., Sander, J. and Xu, X. (1996) A Density Based Algorithm for Discover Clusters in Large Spatial Datasets with Noise. Proceedings of International Conference on Knowledge Discovery and Data Mining, 226-231.

Guha, S., Rastogi, R. and Shim. K, (2001) Cure: An Efficient Algorithm for Large Databses, Information Systems, 26, 35-58.

Kaggle. 2011. Give Me Some Kredit. https://www.kaggle.com/c/GiveMeSomeCredit/ data diunduh 26 Juli 2016.

Laeli, S. (2014). Analisis cluster dengan average linkage method dan ward's method untuk data responden nasabah Asuransi Jiwa Unit Link (Doctoral dissertation, UNY).

Maimon, O., & Rokach, L. (Eds.). (2005). Data mining and knowledge discovery handbook (Vol. 2). New York: Springer.

Mumtaz, K. and Duraiswamy, K., (2010). An analysis on density based clustering of multi dimensional spatial data. Indian Journal of Computer Science and Engineering, 1(1), pp.8-12.

Nagpal, P. B., & Mann, P. A. (2011). Comparative study of density based clustering algorithms. International Journal of Computer Applications, 27(11), 44-47.

Tan, M. Steinbach, and V. Kumar, (2005) Introduction to Data Mining, (First Edition). Boston, MA, USA: Addison-Wesley Longman Publishing CO., Inc.

Yan, B., and Deng, G. (2015). Improved Clustering Algorithm Based on Density-Isoline. Open Journal of Statistics, 5(04), 303-310.




DOI: http://dx.doi.org/10.29313/.v0i0.3796

Flag Counter     Â