Laboratory Clustering using K-Means, K-Medoids, and Model-Based Clustering
| Main Authors: | Qona'ah, Niswatul, Devi, Alvita Rachma, Dana, I Made Gde Meranggi |
|---|---|
| Format: | Article info application/pdf Journal |
| Bahasa: | eng |
| Terbitan: |
Universitas Sebelas Maret
, 2020
|
| Online Access: |
https://jurnal.uns.ac.id/ijas/article/view/40823 https://jurnal.uns.ac.id/ijas/article/view/40823/27728 |
Daftar Isi:
- Institut Teknologi Sepuluh Nopember (ITS) is an institute which has about 100 laboratories to support some academic activity like teaching, research and society service. This study is clustering the laboratory in ITS based on the productivity of laboratory in carrying out its function. The methods used to group laboratory are K-Means, K-Medoids, and Model-Based Clustering. K-Means and K-Medoids are non-hierarchy clustering method that the number of cluster can be given at first. The difference of them is datapoints that selected by K-Medoids as centers (medoids or exemplars) and mean for K-Means. Whereas, Model-Based Clustering is a clustering method that takes into account statistical models. This statistical method is quite developed and considered to have advantages over other classical method. Comparison of each cluster method using Integrated Convergent Divergent Random (ICDR). The best method based on ICDR is Model-Based Clustering.Keywords : K-Means, K-Medoids, Laboratory, Model-Based Clustering