Claim Missing Document
Check
Articles

Found 1 Documents
Search

COMPARISON BETWEEN BICLUSTERING AND CLUSTER-BIPLOT RESULTS OF REGENCIES/CITIES IN JAVA BASED ON PEOPLE’S WELFARE INDICATORS Widyaningsih, Yekti; Nisa, Alfia Choirun
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 19 No 2 (2025): BAREKENG: Journal of Mathematics and Its Application
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol19iss2pp1009-1022

Abstract

The success of a country's development can be known from the well-being of its people. Improving the welfare of the population is the main goal of the development activities carried out by the government. To ensure that development is effective and targeted, grouping is needed to understand the characteristics of the region. This study discusses the grouping of regencies/cities in Java Island based on the people's welfare indicators in 2022. The measured welfare is material well-being. Variables used in this study are the percentage of the poor population, GDP per capita at current prices, average length of schooling, expected length of schooling, percentage of per capita expenditure on food, open unemployment rate, population, population density, and life expectancy. There are two approaches used in grouping regencies/cities along with their variables. The first approach is to simultaneously group regencies/cities and their variables using Plaid Model biclustering. The second approach is to group regencies/cities using the Ward clustering method followed by the biplot method. This study aims to compare the results of these two approaches, namely the biclustering and cluster-biplot results, on data from 119 regencies/cities in Java Island in 2022 based on people's welfare indicators. Based on the results of this study, the number of groups from each approach is 2, with group 1 being more prosperous than group 2. Judging from the standard deviation values, the Plaid Model biclustering result groups have lower standard deviation values than the cluster-biplot result groups. Therefore, in general the first approach produces better groups as they are more homogeneous than the second approach.