ON A CLUSTERING OPTIMIZATION WITH GENETIC ALGORITHM OF FUZZY C-MEANS AND FUZZY GUSTAFSON-KESSEL (CASE STUDY: FISHER’S IRIS)

Novitasari, Kiki Indah and Purba, Lasman Parulian and Saputra, Wahyu S. J. (2017) ON A CLUSTERING OPTIMIZATION WITH GENETIC ALGORITHM OF FUZZY C-MEANS AND FUZZY GUSTAFSON-KESSEL (CASE STUDY: FISHER’S IRIS). In: Prosiding Seminar Nasional Sains, Rekayasa & Teknologi UPH - 2017. SNSRT-2017, 2 . Faculty of Science and Technology, Karawachi, Jakarta, I-51-I-55. ISBN 978-979-1053-06-8

	Text 4_o_Prosiding FaST 2017_Karya ON A CLUSTERING OPTIMIZATION (1).pdf Download (569kB)
	Text o_HASIL SIMILARITY-ON A CLUSTERING OPTIMIZATION WITH GENETIC ALGORITHM OF FUZZY C-MEANS AND FUZZY GUSTAFSON-KESSEL (CASE STUDY_ FISHER’S IR.pdf Download (900kB)
	Text o_Peer review_ON A CLUSTERING OPTIMIZATION WITH GENETIC ALGORITHM OF FUZZY C-MEANS AND FUZZY GUSTAFSON-.pdf Download (1MB)

Official URL: http://fast.uph.edu/semnas/2017/84.html

Abstract

A Fuzzy Gustafson-Kessel (FGK) is one of clustering method using adaptive norm-distance to detect the shape of each data cluster. This algorithm is a development of Fuzzy C-means (FCM) that the result remains local minimum solution, thus the genetic algorithm (GA) approach is used to solve that problem. Then the clustering process uses MATLAB R2012a using FGK Algorithm with GA optimization. This optimization process from FGK clustering using GA is started by inputting the tested data, Fisher’s Iris. Resulting the matrix of cluster center from FGK process, then the evolution must be done using GA to make matrix of cluster center more optimal. Based on the test, it can be summarized that the optimization from FGK clustering of Fisher’s Iris data set using GA will be better by minimizing objective function. Thus, the objective function value of FGK-GA resulted is smaller than FGK in all tested cluster values. Based on the FGK-GA classification rate 90.31% is more than the average value of FGK classification rate 90%. The test showed that the best cluster is 3 and this value is similar to Fisher’s Iris data set classified in 3 classes.

Item Type:	Book Section
Uncontrolled Keywords:	Clustering, Fuzzy Gustafson-Kessel, Genetic Algorithm, Fisher’s Iris, Classification Rate, Objective function
Subjects:	H Social Sciences > HD Industries. Land use. Labor
Divisions:	Fakultas Teknik > Prodi Teknik Industri
Depositing User:	Lasman Parulian Purba
Date Deposited:	04 Nov 2021 07:41
Last Modified:	03 Oct 2022 04:16
URI:	http://repositori.ukdc.ac.id/id/eprint/760

Actions (login required)

View Item