SEGMENTASI DAN PENENTUAN KLASTER TERBAIK MENGGUNAKAN KRITERIA ELBOW DAN METODE HIRARKI K-MEANS DENGAN PROSES SKORING; SEGMENTATION AND DETERMINE THE BEST CLUSTER USING ELBOW CRITERION AND HIERARCHICAL K-MEANS METHOD WITH SCORING PROCESS

Bakti, Darmawan Ratdya, Yunita Wulan Sari

2015 | Skripsi | FMIPA

Methods to perform clustering process is divided into hierarchical method and non-hierarchical method. Non-hierarchical method or k-means clustering is grouping data into k clusters. This process begins with determine number of cluster, data with the same characteristics will be included in the same group while the data which have different characteristics will be entered into different groups. In the process of using the k-means clustering of the number of k is determined in advance, one of the methods used in this analysis is using the elbow criterion. In addition another clustering technique is hierarichical method, in this research using centroid linkage analysis, where the agglomeration method used in this process with each of the data that is formed into a cluster and will be determined central value/centroid to cluster desired. In this paper using hierarchical k-means method, This method combines hierarchical method and non-hierarchical method, hierarichical process is used to determine the initial centroid for non-hierarchical process and obtain optimal cluster.

Kata Kunci : k-means; hierarichal algorithm; elbow criterion; clustering; initial centroid determiation