JAMBURA JOURNAL OF PROBABILITY AND STATISTICS
Vol 1, No 2 (2020): Jambura Journal of Probability and Statictics

PENGGUNAAN SELF ORGANIZING MAP DALAM PENGELOMPOKAN TINGKAT KESEJAHTERAAN MASYARAKAT

IRWAN IRWAN ([SINTA ID: 6652752] Department of Mathematics, Universitas Negeri Makassar)
ASTRI YUNI HASHARI (Department of Mathematics, Universitas Negeri Makassar)
HISYAM IHSAN (Department of Mathematics, Universitas Negeri Makassar)
AHMAD ZAKI (Department of Mathematics, Universitas Negeri Makassar)



Article Info

Publish Date
22 Sep 2020

Abstract

Self Organizing Map (SOM) is one of the topology forms of Unsupervised Neural Network where in the learning process does not require output target. Clusters in this research consist of one or more regency/city areas that have certain characteristics based on the variables. Each cluster had to be validated by using the Davies Bouldin Index value to get the best cluster formation from the SOM algorithm learning process. The best cluster model is the cluster model that has the smallest Davies Bouldin Index value. This research used 30 variables that refer to the key statistics of South Sulawesi Province People's Prosperity in 2018 by BPS of South Sulawesi Province. In this research, four cluster formation models were formed which began by forming 2 cluster model to form 5 cluster. Based on the Davies Bouldin Index value, it was found that the  5 cluster model have minimum value of 0.17.

Copyrights © 2020






Journal Info

Abbrev

jps

Publisher

Subject

Computer Science & IT Decision Sciences, Operations Research & Management Environmental Science Social Sciences

Description

Probability Theory Mathematical Statistics Computational Statistics Stochastic Processes Financial Statistics Bayesian Analysis Survival Analysis Time Series Analysis Neural Network Another field which is related to statistics and the applications Another field which is related to Probability and ...