STKIP PGRI BLITAR

COMPARATIVE ANALYSIS OF DISADVANTAGED AREAS IN REGENCIES/CITIES IN EASTERN INDONESIA IN 2021 USING THE K-MEANS CLUSTERING, K-MEDOIDS CLUSTERING, AND FUZZY C-MEANS CLUSTERING METHODS

Bayu Aji , Bachtiar and Udin , Erawanto (2024) COMPARATIVE ANALYSIS OF DISADVANTAGED AREAS IN REGENCIES/CITIES IN EASTERN INDONESIA IN 2021 USING THE K-MEANS CLUSTERING, K-MEDOIDS CLUSTERING, AND FUZZY C-MEANS CLUSTERING METHODS. Cakrawala Pendidikan, 28 (1). pp. 1-17. ISSN 1410-9883

[img]PDF
590Kb

Official URL: http://id.stkippgri-blitar.ac.id/digilib

Abstract

Abstrak: Tujuan dari penelitian ini adalah untuk membandingkan penggunaan metode K-Means Clustering, K-Medoids Clustering dan Fuzzy C-Means Clustering dalam klasifikasi daerah tertinggal kabupaten/kota di Indonesia Timur. Data yang digunakan dalam penelitian ini adalah data sekunder Statistik Keuangan Pemerintah Kabupaten/Kota edisi 2020/2021 di website BPS dengan menggunakan unit observasi Sulawesi Utara, Sulawesi Selatan, Maluku, Maluku Utara, NTB dan unit observasi provinsi kabupaten/kota. Daerah Beda Persentase Penduduk Miskin di Papua Barat, Dana Alokasi Khusus Daerah, Angka Harapan Hidup, Jumlah Puskesmas. Hasil penelitian ini menunjukkan bahwa clustering singular dengan menggunakan algoritma k-means lebih optimal dibandingkan dengan clustering dengan algoritma k-medoids atau fuzzy c-means untuk indikator daerah tertinggal di Indonesia bagian timur. Hasil klasterisasi k-means terhadap 176 daerah di Indonesia bagian timur berdasarkan indikator daerah tertinggal terbagi menjadi 3 klaster yaitu klaster sangat tertinggal dan tidak tertinggal. Secara total, terdapat 91 kabupaten/kota yang tidak tertinggal jauh, 48 kabupaten/kota tertinggal sedang, dan 37 kabupaten/kota tertinggal. Kata kunci: Analisis Komparatif, Daerah Tertinggal, Indonesia Bagian Timur, K-Means Clustering, K-Medoids Clustering, Metode Fuzzy C-Means Clustering Abstract: The purpose of this study is to compare the use of K-Means Clustering, K-Medoids Clustering and Fuzzy C-Means Clustering methods in the classification of districts/urban underdeveloped areas in Eastern Indonesia. The data used in this study is secondary data from the Regency/City Government Financial Statistics 2020/2021 edition on the BPS website using North Sulawesi, South Sulawesi, Maluku, North Maluku, NTB and provincial districts/cities observation units. area Different percentage of poor people in West Papua, Regional Special Allocation Fund, Life Expectancy, number of health centers. The results of this study show that singular clustering using k-means algorithms is more optimal than clustering with k-medoids or fuzzy c-means algorithms for indicators of disadvantaged areas in eastern Indonesia. The k-means clustering results of 176 regions in eastern Indonesia, based on the indicators of underdeveloped areas, are divided into 3 clusters, namely very underdeveloped clusters and not backward. In total, 91 districts are not far behind, 48 are moderately behind and 37 are behind. Keywords: Comparative Analysis, Disadvantaged Areas, Eastern Indonesia, K-Means Clustering, K-Medoids Clustering, Fuzzy C-Means Clustering Methods

Item Type:Article
Subjects:L Education > L Education (General)
Divisions:Faculty of Law, Arts and Social Sciences > School of Education
ID Code:1013
Deposited By:FERI HUDA, M.Pd
Deposited On:21 Jul 2024 07:12
Last Modified:21 Jul 2024 07:12

Repository Staff Only: item control page