esc

Type at least two characters to start searching.

↑↓ select↵ openesc close

Share

Export Citation

APA
MLA
Chicago
Harvard
Vancouver
BIBTEX
RIS
Universitas Hasanuddin
Research output:Contribution to journal›Article›peer-review

Evaluation of the Clustering Results in Dermatology Data Using the Silhouette Coefficient

Adiba F.

Aip Conference Proceedings

Published: 2024Citations: 3

Abstract

Clustering is a data analysis method widely used to obtain information from a set of data. One of the obstacles in using the clustering method is determining the number of clusters and various other influential variables. The clustering method used in this research is clustering self-organizing maps (SOM) using skin disease data. This study aims to determine the optimum variables to define the clustering process variables by evaluating clustering results. This study evaluates the clustering results using internal standards using the silhouette coefficient method. Silhouette coefficient is an evaluation method that combines cohesion and separation methods—clustering with SOM in this study by testing several values to obtain optimal variable values. The variables tested are the number of clusters, epochs, and learning rate. The results of the clustering evaluation using the silhouette coefficient method are 0.316 for each cluster variable, the number of groups K = 4, epoch = 50, and learning rate = 0.5.

Access to Document

10.1063/5.0164548

Other files and links

Fingerprint

SilhouetteSciences
Cluster analysisSciences
Computer scienceSciences
Artificial intelligenceSciences
Computer graphics (images)Sciences
DermatologySciences
Computer visionSciences
Pattern recognition (psychology)Sciences
MedicineSciences