Penerapan Algoritme K-Prototypes untuk Pengelompokkan Desa-Desa di Provinsi Jawa Barat Berdasarkan Indikator Indeks Desa Membangun Tahun 2020

Mayang Ganmanah, Abdul Kudus

Abstract


Abstract. The grouping of villages in West Java needs to be done as material for planning and evaluating government program targets. The aim is to determine the status of village development based on the indicators of the developing village index and so that the policy programs carried out by the government are more focused according to the characteristics of the results of village grouping. Thus, it is necessary to conduct an analysis related to the clustering of the indicators of the developing village index. The analysis that can be used to group a village based on its characteristics is cluster analysis. The K-Prototypes algorithm is a cluster analysis method for large data with mixed-type data. A total of 3698 villages from 18 regencies and 9 cities were used as observation units. The grouping of villages in West Java Province based on the index of developing villages in 2020 using the K-prototypes algorithm produces 5 clusters. In the 1st cluster there are 54 villages, the 2nd cluster there are 1273 villages, the 3rd cluster there are 1272 villages In the 4th cluster there are 464 villages and in the 5th cluster there are 635 villages. Where the 5th cluster is the best cluster among the other 5 clusters.

Keywords: Clustering, K-Prototypes, Village Index Build.

Abstrak. Pengelompokkan desa di Jawa Barat perlu dilakukan sebagai bahan perencanaan dan evaluasi sasaran program pemerintah. Tujuannya untuk mengetahui penetapan status perkembangan desa berdasarkan indikator indeks desa membangun dan agar program kebijakan yang dilakukan pemerintah lebih terarah sesuai karakteristik hasil dari pengelompokkan desa. Dengan demikian perlu dilakukan analisis terkait klasterisasi terhadap indikator indeks desa membangun. Analisis yang dapat digunakan untuk mengelompokkan suatu desa berdasarkan karakteristik-karakteristik yang dimilikinya adalah analisis cluster. Algoritme K-Prototypes merupakan salah satu metode analisis cluster pada data berukuran besar dengan data bertipe campuran. Sebanyak 3698 desa dari 18 Kabupaten dan 9 Kota digunakan sebagai unit pengamatan. Pengelompokan desa-desa di Provinsi Jawa Barat berdasarkan indikator indeks desa membangun tahun 2020 dengan menggunakan algoritme K-prototypes menghasilkan 5 cluster. Pada cluster ke-1 terdapat 54 desa, cluster ke-2 terdapat 1273 desa, cluster ke-3 terdapat 1272 desa, cluster ke-4 terdapat 464 desa dan pada cluster ke-5 terdapat 635 desa. Dimana cluster ke-5 merupakan cluster terbaik diantara 5 cluster lainnya.

Kata Kunci: Clustering,K-Prototypes, Indeks Desa Membangun.


Keywords


Clustering, K-Prototypes, Indeks Desa Membangun.

Full Text:

PDF

References


Bunkers, M., Miller, J., & DeGaetano, A. (1996). Definition of climate regions in the northen plains using an objective cluster modification technique. Journal of Climate, 130-146.

Gan, G., MA, C., & Wu, J. (2007). Data Clustering Theory, Algorithms, sand Applications. Virginia (US): American Statistical Association (ASA).

Hair, J., Black, W., Babin, J., & Anderson, R. (2010). Multivariate Data Analysis "7th ed". New Jersey(US): Pearson Prentice Hall.

Hajarisman, N. (2019). Statistika Multivariat: Analisis Klaster. Bandung: NH Press.

Huang, Z. (1998). Extensions to the k-Means Algorithm for Clustering Large data Sets with Categorical Values . Data Mining and Knowledge Discovery, 2833-304.

Kader, G., & Perry, M. (2007). Variability for categorical variables. Journal of Statistics, 15(2): 1-16.

Kementerian Desa, PDT dan Transmigrasi. (2019). Status IDM Indeks Desa Membangun Provinsi Kabupaten Kecamatan Tahun 2019. Jakarta Selatan: Direktorat Jenderal Pembangunan dan Pemberdayaan Masyarakat Desa.

Okada, T. (1999). Sum of Squares Decompositionfor Categorical Data. Kwansei Gakuin Studies in Computer Science, 14:1-6.

Rencher, A. C. (2007). Methods of Multivariate Analysis Second Edition. Canada (US): Johm Wiley & Sons.

Yulianto Anggi Priliani, Darwis Sutawanir. (2021). Penerapan Metode K-Nearest Neighbors (kNN) pada Bearing. Jurnal Riset Statistika, 1(1), 10-18.




DOI: http://dx.doi.org/10.29313/.v0i0.28974

Flag Counter