CLUSTERING RURAL DEVELOPMENT TYPOLOGY IN EAST JAVA PROVINCE USING LATENT CLASS ANALYSIS

Desy Setiawati, . and Aji Hamim Wigena, . and Bagus Sartono, . (2015) CLUSTERING RURAL DEVELOPMENT TYPOLOGY IN EAST JAVA PROVINCE USING LATENT CLASS ANALYSIS. Proceeding of International Conference On Research, Implementation And Education Of Mathematics And Sciences 2015 (ICRIEMS 2015), Yogyakarta State University, 17-19 May 2015. ISSN 978-979-96880-8-8

[img]
Preview
Text
M -23.pdf

Download (212kB) | Preview
Official URL: http://fmipa.uny.ac.id

Abstract

To deliver the sustained and equitable regional development in Indonesia, the government must understand the characteristics of each region based on its area features, therefore, the classification of rural area must be handled to increase the precision of the development program. Since the rural area has their own specific characteristics that may lead to its rural level, the classification must ensure that the development policy fit in each area. In this paper, we try to classify the typology of rural development that measured based on the rural potential characteristics, education, and socioeconomic. We select villages in East Java province as a scope of research area since East Java was well-known as a center of agricultural in Java, however, in 2011-2014, according to BPS, the poverty rate put east Java in 15th position in the national poverty rate. The classification uses latent class analysis, which models the data into particular statistical distribution to identify immeasurable cluster membership between subjects with observed categorical or continuous variables. The method was able to handle overlapping model data by setting different characteristics, and the modeling results can be tested its accuracy level. Expectation Maximization (EM) algorithm is used to estimate parameters of the latent class model. The research uses PODES 2011 dataset which contains characteristics and facilities information of 8502 villages. The result showed that using latent class analysis generates five clusters of rural area development, while the current classification from Ministry of Home Affairs only uses three typologies of rural development. The research result was able to give more detail additional information of current three classifications by dividing its typology into several detail typology classifications. Key words: Latent Class Analysis, Maximum Likelihood, Expectation Maximization Algorithm, Rural Development Typology

Item Type: Article
Subjects: Prosiding > ICRIEMS 2015 > Mathematics & Mathematics Education
Divisions: Fakultas Matematika dan Ilmu Pengetahuan Alam > Jurusan Pendidikan Matematika > Matematika
Depositing User: Administrator
Date Deposited: 15 Jul 2015 12:38
Last Modified: 15 Jul 2015 12:38
URI: http://eprints.uny.ac.id/id/eprint/23632

Actions (login required)

View Item View Item