Analisis Pengelompokan Provinsi di Indonesia Berdasarkan Indikator Stunting dengan Metode K-Means dan DBSCAN

Ramadhana, Ikrar Setya (2025) Analisis Pengelompokan Provinsi di Indonesia Berdasarkan Indikator Stunting dengan Metode K-Means dan DBSCAN. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 2043211047-Undergraduate_Thesis.pdf] Text
2043211047-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (3MB) | Request a copy

Abstract

Masalah stunting di Indonesia perlu mendapatkan perhatian lebih dari pemerintah karena angka prevelensi stunting di Indonesia sangat besar, hal ini dibuktikan dengan data dari WHO tahun 2020 yang menempatkan Indonesia di urutan ke-4 di dunia dan urutan ke-2 di Asia Tenggara sebagai negara dengan prevelensi stunting terbesar. Program penurunan stunting pemerintah sudah terlihat baik secara angka namun belum merata. Cara penelitian ini membantu pemerintah yaitu dengan melakukan pengelompokan wilayah provinsi di Indonesia berdasarkan indikator stunting dengan menggunakan metode clustering. Tujuan penelitian ini adalah membandingkan metode pengelompokan wilayah di Indonesia antara metode K-means dan DBSCAN untuk penanganan kasus balita stunting. Berdasarkan hasil analisis metode K- means memiliki nilai DBI yang lebih rendah, artinya metode K-means lebih efektif untuk pengelompokan wilayah berdasarkan Indikator stunting. Dari metode ini terbentuk 6 cluster dengan cluster 1 terdiri dari 7 wilayah, cluster 2 terdiri dari 8 wilayah, cluster 3 terdiri dari 1 wilayah, cluster 4 terdiri dari 14 wilayah, cluster 5 terdiri dari 2 wilayah, dan cluster 6 terdiri dari 6 wilayah, Dengan faktor – faktor yang memengaruhi secara signifikan terhadap angka prevalensi stunting yaitu persentase status gizi normal.
===================================================================================================================================
The problem of stunting in Indonesia needs to get more attention from the government because the stunting prevalence rate in Indonesia is very large, this is evidenced by data from WHO in 2020 which places Indonesia in 4th place in the world and 2nd place in Southeast Asia as the country with the largest stunting prevalence. The government's stunting reduction program has looked good in numbers but has not been evenly distributed. The way this research helps the government is by clustering provincial areas in Indonesia based on stunting indicators using the clustering method. The purpose of this research is to compare regional clustering methods in Indonesia between the K-means and DBSCAN methods for handling cases of stunting toddlers. Based on the results of the analysis, the K-means method has a lower DBI value, meaning that the K-means method is more effective for clustering areas based on stunting indicators. From this method, 6 clusters are formed with cluster 1 consisting of 7 regions, cluster 2 consisting of 8 regions, cluster 3 consisting of 1 region, cluster 4 consisting of 14 regions, cluster 5 consisting of 2 regions, and cluster 6 consisting of 6 regions, with factors that significantly affect the stunting prevalence rate, namely the percentage of normal nutritional status.

Item Type: Thesis (Other)
Uncontrolled Keywords: Density Based Spatial Clustering of Application with Noise, Klaster K-Means, dan Stunting. Density Based Spatial Clustering of Application with Noise, K-Means Clustering, and Stunting.
Subjects: Q Science > QA Mathematics > QA278.55 Cluster analysis
Divisions: Faculty of Vocational > 49501-Business Statistics
Depositing User: Ikrar Setya Ramadhana
Date Deposited: 21 Jul 2025 08:43
Last Modified: 21 Jul 2025 08:43
URI: http://repository.its.ac.id/id/eprint/120325

Actions (login required)

View Item View Item