ANALISA PENENTUAN JUMLAH CLUSTER TERBAIK PADA METODE K-MEANS CLUSTERING

Abstract

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