Open AccessAnalysis of Euclidean Distance and Manhattan Distance in the K-Means Algorithm for Variations Number of Centroid K
Author(s) -
R. Suwanda,
Z. Syahputra,
Elviawaty Muisa Zamzami
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1566/1/012058
Subject(s) - centroid , euclidean distance , cluster analysis , distance matrix , partition (number theory) , mathematics , k medians clustering , cluster (spacecraft) , euclidean geometry , mahalanobis distance , minkowski distance , algorithm , combinatorics , computer science , pattern recognition (psychology) , statistics , artificial intelligence , fuzzy clustering , cure data clustering algorithm , geometry , programming language
K-Means is a clustering algorithm based on a partition where the data only entered into one K cluster, the algorithm determines the number group in the beginning and defines the K centroid. The initial determination of the cluster center is very influential on the results of the clustering process in determining the quality of grouping. Better clustering results are often obtained after several attempts. The manhattan distance matrix method has better performance than the euclidean distance method. The author making the result of conducted testing with variations in the number of centroids (K) with a value of 2,3,4,5,6,7,8,9 and the authors having conclusions where the number of centroids 3 and 4 have a better iteration of values than the number of centroids that increasingly high and low based on the iris dataset.