Comparison of hierarchical clustering methods (case study: data on poverty influence in North Sulawesi)

3rd Indonesian Operations Research Association - International Conference on Operations Research 2018

计算机科学

Mongi, C.E.^1 ; Langi, Y.A.R.^1 ; Montolalu, C.E.J.C.^1 ; Nainggolan, N.^1

Department of Mathematics, Faculty of Mathematics and Natural Science, University of Sam Ratulangi, Manado, Indonesia^1

关键词: Agglomerative hierarchical clustering;    Average linkage;    Centroid method;    Hierarchical clustering methods;    Large data;    Ward method;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/567/1/012048/pdf
DOI  :  10.1088/1757-899X/567/1/012048

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Grouping of Large data has been carried out in various fields. One method for grouping is cluster analysis where this method consists of hierarchy and non-hierarchy method. The aim of this study was to compare the use of cluster analysis on aspects of the causes of poverty data. The method used is agglomerative hierarchical clustering, that is, the average linkage, centroid methods and ward methods. The results obtained are compared with the RMSSTD value and the smallest value is the ward method with a value of 2.0937. So the ward method is good for this case.