Open AccessFuzzy c-means and gath-geva methods in clustering districts based on human development index (hdi) in south sulawesi
Author(s) -
Suwardi Annas,
Sukri Nyompa,
R. Arisandi,
Muhammad Nusrang,
Eka Sriwahyuni
Publication year - 2019
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 - 10.1088/1742-6596/1317/1/012014
Subject(s) - fuzzy logic , cluster analysis , human development index , fuzzy clustering , data mining , computer science , artificial intelligence , flame clustering , mathematics , human development (humanity) , canopy clustering algorithm , political science , law
District grouping in South Sulawesi based on the Human Development Index (HDI) indicators needs to be done as a material for planning and evaluating the targets of government work programs. This grouping is based on dominant indicators of the high and low HDI. The value of the HDI indicator needs to be considered so that the achievement of each indicator is known. Statistical analysis that can be used to group districts that have similarities is cluster analysis. The method that is currently developing is fuzzy clustering analysis, which classifies objects using certain membership degrees. Fuzzy clustering algorithm that can be used is Fuzzy C-means (FCM). Another method of fuzzy clustering analysis developed further is Gath Geva (GG), which is able to detect groups with different forms. In this study, the fuzzy clustering process on the FCM and GG methods with the same parameters and shows that the GG method is better than the FCM method. This conclusion is based on a total of 1000 iterations. The GG method gives an objective function value smaller than FCM, besides it gives a faster- conferencing iteration result.