Regency grouping in East Java based on Variable Type of Agriculture uses Hybrid Hierarchical Clustering Via Mutual Cluster Method
Abstract East Java Province is one of the provinces that has the largest agricultural resources in Indonesia. The Government of East Java needs to produce superior commodities in each region. This study aims to group districts in East Java Province based on variable types of agriculture with the hyb...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
UIN Syarif Hidayatullah
2020-03-01
|
Series: | InPrime |
Subjects: | |
Online Access: | https://journal.uinjkt.ac.id/index.php/inprime/article/view/14167 |
Summary: | Abstract
East Java Province is one of the provinces that has the largest agricultural resources in Indonesia. The Government of East Java needs to produce superior commodities in each region. This study aims to group districts in East Java Province based on variable types of agriculture with the hybrid hierarchical clustering via mutual cluster method that combines the merging of bottom-up clustering advantages and top-down clustering advantages. Mutual cluster is a grouping with the largest distance between small groups of the shortest distance for each point outside the group. In this research, the calculation uses Euclidean distance. The data used in this study are from the East Java Central Statistics Agency (BPS) in 2017. The division calculation is obtained by finding the minimum (standard deviation of intra cluster) value and the maximum (standard deviation of inter clusters) value and using the analysis of variance calculation. The grouping results obtained were nine groups with value of 725.934, value of 1.475.978 and value of 7,908.
Keywords: agriculture; Hybrid Hierarchical Clustering; mutual cluster; Euclidean distance; analysis of variance.
Abstrak
Provinsi Jawa Timur merupakan salah satu provinsi yang memiliki sumber daya pertanian terbesar di Indonesia. Pemerintah Jawa Timur perlu mengembangkan komoditi unggulan di tiap daerah di Jawa Timur. Penelitian ini bertujuan untuk mengelompokkan kabupaten di Provinsi Jawa Timur berdasarkan variabel jenis pertanian dengan metode hybrid hierarchical clustering via mutual cluster yaitu menggabungkan kelebihan bottom-up clustering dan kelebihan top-down clustering. Mutual cluster yakni pengelompokkan dengan jarak terbesar antara bagian dalam kelompok yang kecil dari jarak yang terpendek kepada tiap titik di luar kelompok. Dalam penelitian ini, perhitungan jarak menggunakan jarak Euclidean. Data yang digunakan dalam penelitian ini dari Badan Pusat Statistik Jawa Timur tahun 2017. Perhitungan pembagian didapat dengan mencari nilai (simpangan baku dalam klaster) yang minimal dan nilai (simpangan baku antar klaster) yang maksimal, serta digunakan perhitungan analyze of varians. Hasil pengelompokkan yang diperoleh didapatkan sebanyak sembilan kelompok dengan nilai sebesar 725.934, nilai sebesar 1.475.978 dan nilai sebesar 7,908.
Kata Kunci: pertanian; Hybrid Hierarchical Clustering; mutual cluster; jarak Euclid; analisis variansi. |
---|---|
ISSN: | 2686-5335 2716-2478 |