Implementation of Minkowski-Chebyshev Distance in Fuzzy Subtractive Clustering

Annisa Eka Haryati (Master in Mathematics Education, Universitas Ahmad Dahlan, Yogyakarta, Indoneisa)
Sugiyarto Surono (Mathematics Faculty of Applied Science and Technology Universitas Ahmad Dahlan Yogyakarta, Indonesia)
Suparman Suparman (Master in Mathematics Education, Universitas Ahmad Dahlan Yogyakarta, Indoneisa)

EKSAKTA: Journal of Sciences and Data Analysis
VOLUME 2, ISSUE 2, August 2021

Clustering is a method of the grouping which is done by looking at the similarities between data in a data set. Fuzzy clustering is a clustering method that uses fuzzy set membership values as the basis for grouping data. Fuzzy Subtractive Clutering (FSC) is a fuzzy clustering method where the number of clusters to be formed is unknown. The concept of FSC is to determine the highest data density and the data with the most number of neighbors will be selected as the center of the cluster. Thus, the size of the proximity or distance between points is needed to determine the members of each cluster. The distance used in this study is a combination of the Minkowski and Chebyshev distances. The number of clusters formed will be evaluated using the Partition Coefficient (PC) value where the highest PC value indicates the best number of clusters. The results obtained indicate that the best clusters are three clusters with a PC value of 0.7422