Proceeding SENDI_U
2015: SEMINAR NASIONAL MULTI DISIPLIN DAN CALL FOR PAPERS

ANALISA PENENTUAN JUMLAH CLUSTER TERBAIK PADA METODE K-MEANS CLUSTERING

Eka Merliana, Ni Putu (Unknown)
., Ernawati (Unknown)
Santoso, Alb. Joko (Unknown)



Article Info

Publish Date
18 Aug 2015

Abstract

Clustering is a technique used to analyze data either in machine learning, data mining, patternrecognition, image analysis and bioinformatics. So as to produce useful information need for ananalysis of data using clustering process because data has a lot of variety and quantity. In thiscase the researchers will use the K-Means method in which these methods into an efficient andeffective algorithms to process data with the variety and number of lots. K-means algorithm hasa problem in determining the best number of clusters. So in this paper the researchers willconduct research to search for the best number of clusters in K-means method. There are manyways to determine this, one of them with methods Elebow. The determination of these methodsseen from the graph SSE (Sum Square Error) of some number of clusters. Results from this studywill be the basis for determining the number clusters in the process clustering with K-Meansmethod in a case study, and this case study will be conducted at the institute STAHN (SekolahTinggi Agama Hindu Negeri) Tampung Penyang Palangkaraya.Keywords: Clustering, K-Means, Method Elbow, SSE (Sum Square Error)

Copyrights © 2015