ANALISIS CLUSTERING BERDASARKAN FREKUENSI KEJADIAN DAN DAMPAK BENCANA HIDROMETEOROLOGI DI INDONESIA MENGGUNAKAN SELF ORGANIZING MAPS DAN K-MEDOIDS

GLORIA KRISTIN NATALIA SITANGGANG, . (2026) ANALISIS CLUSTERING BERDASARKAN FREKUENSI KEJADIAN DAN DAMPAK BENCANA HIDROMETEOROLOGI DI INDONESIA MENGGUNAKAN SELF ORGANIZING MAPS DAN K-MEDOIDS. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text 1. File COVER.pdf Download (1MB)
	Text 2. File BAB I.pdf Download (288kB)
	Text 3. File BAB II.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text 4. File BAB III.pdf Restricted to Registered users only Download (527kB) | Request a copy
	Text 5. File BAB IV.pdf Restricted to Registered users only Download (805kB) | Request a copy
	Text 6. File BAB V.pdf Restricted to Registered users only Download (254kB) | Request a copy
	Text 7. File DAFTAR PUSTAKA.pdf Download (238kB)
	Text 8. File Lampiran.pdf Restricted to Registered users only Download (415kB) | Request a copy

Abstract

Bencana hidrometeorologi merupakan jenis bencana yang paling dominan terjadi di Indonesia dan menimbulkan dampak yang berbeda pada setiap provinsi. Perbedaan karakteristik tersebut menunjukkan perlunya pengelompokan provinsi agar pola kemiripan berdasarkan frekuensi kejadian dan dampak bencana dapat diidentifikasi. Penelitian ini bertujuan untuk mengidentifikasi karakteristik, pola kemiripan, serta mengelompokkan provinsi di Indonesia berdasarkan frekuensi kejadian dan dampak bencana hidrometeorologi menggunakan pendekatan two-stage clustering melalui Self Organizing Maps (SOM) dan K-Medoids. Data yang digunakan merupakan data sekunder Badan Nasional Penanggulangan Bencana (BNPB) tahun 2025 yang mencakup 38 provinsi dengan tujuh variabel, yaitu jumlah kejadian, jumlah fasilitas umum rusak, jumlah rumah terendam, jumlah rumah rusak, jumlah orang terluka, jumlah meninggal, dan jumlah orang hilang. Hasil penelitian menunjukkan bahwa visualisasi SOM mampu menggambarkan karakteristik dan pola kemiripan antarprovinsi berdasarkan kombinasi ketujuh variabel tersebut. Selanjutnya, pendekatan two-stage clustering menghasilkan empat kelompok provinsi dengan karakteristik frekuensi kejadian dan dampak bencana hidrometeorologi yang berbeda. Hasil pengelompokan ini diharapkan dapat menjadi salah satu referensi dalam memahami karakteristik bencana hidrometeorologi antarprovinsi di Indonesia. Kata kunci: Self Organizing Maps, K-Medoids, Two-stage Clustering, Bencana Hidrometeorologi, Clustering. ***** Hydrometeorological disasters are the most dominant type of disaster in Indonesia and have different impacts on each province. These differences in characteristics indicate the need for provincial groupings to identify similarity patterns based on the frequency of occurrence and disaster impacts. This study aims to identify the characteristics, similarity patterns, and group provinces in Indonesia based on the frequency of occurrence and impact of hydrometeorological disasters using a two-stage clustering approach through Self Organizing Maps (SOM) and K-Medoids. The data used is secondary data from the National Disaster Management Agency (BNPB) in 2025, covering 38 provinces with seven variables: number of incidents, number of damaged public facilities, number of submerged houses, number of damaged houses, number of injured people, number of deaths, and number of missing people. The results of the study indicate that the SOM visualization is able to describe the characteristics and similarity patterns between provinces based on the combination of these seven variables. Furthermore, the two-stage clustering approach produces four groups of provinces with different characteristics of the frequency of occurrence and impact of hydrometeorological disasters. The results of this grouping are expected to serve as a reference in understanding the characteristics of hydrometeorological disasters between provinces in Indonesia. Keywords: Self Organizing Maps, K-Medoids, Two-stage Clustering, Hydrometeorological Disasters, Clustering.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Prof. Dr. Ir. Bagus Sumargo, M.Si. ; 2). Faroh Ladayya, M.Si.
Subjects:	Sains > Statistika
Divisions:	FMIPA > S1 Statistika
Depositing User:	Gloria Kristin Natalia Sitanggang .
Date Deposited:	17 Sep 2026 04:44
Last Modified:	17 Sep 2026 04:44
URI:	http://repository.unj.ac.id/id/eprint/73887

Actions (login required)

View Item