Melola Discovery

← Melola Discovery

Akses terbuka Jurnal SINTA 1

Assessing the impact of training samples overlap and density in random forest for landslide susceptibility mapping: Implications for degraded land management in Bandung Regency, Indonesia

Journal of Degraded and Mining Lands Management · Vol. 12 No. 5 (2025) · hlm. 8933–8956

DOI
10.15243/jdmlm.2025.125.8933
ISSN
2502-2458, 2339-076X
Terbit
2025-10-01
Penerbit
Brawijaya University
Bahasa
eng
Lisensi
creativecommons.org/licenses/by-nc/4.0

Abstrak

Landslide susceptibility mapping is essential for disaster mitigation and land management in degraded mountainous regions. Machine learning algorithms, particularly Random Forest (RF), have been increasingly applied due to their robustness in handling complex, non-linear relationships. However, classification performance is often affected by the quality of training samples, especially when landslide and non-landslide points exhibit spatial overlap. This study investigated how varying densities of fully overlapping samples influence RF performance in Bandung Regency, West Java, Indonesia, an area characterised by steep slopes, rapid land-use change, and post-mining degradation. Balanced datasets ranging from 50 to 700 samples per class were evaluated with hyperparameter tuning. The highest validation accuracy (89%) was achieved with 500 samples at a max_depth of 2, while training accuracy was approximately 10% lower, indicating the algorithm’s difficulty in separating overlapping classes. A more stable trade-off was obtained with 300 samples and a max_depth of 4, suggesting that moderate densities enhance generalisation. To translate these findings into practice, we propose an ensemble zoning and uncertainty mapping framework that integrates multiple model outputs to identify consensus zones for slope stabilisation, vegetation restoration, and adaptive spatial planning. This approach improves the reliability of susceptibility maps and provides actionable insights for managing degraded and landslide-prone landscapes.

Kata kunci

Jaringan kolaborasi para penulis

12 penulis · 27 tautan kepenulisan bersama · 7 lintas organisasi

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • Bandung Institute of Technology
  • Institut Teknologi Nasional Bandung
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Agung Budi HartoBandung Institute of Technology2
Akhmad RiqqiBandung Institute of Technology1
Dini Aprilia NorvyaniBandung Institute of Technology1
Ketut WikantikaBandung Institute of Technology1
Asep SaepulohBandung Institute of Technology1
Satria BijaksanaBandung Institute of Technology1
Ishak Hanafiah IsmullahBandung Institute of Technology1
Widyo Nugroho SulasdiBandung Institute of Technology1

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OAI-PMH (metadata jurnal) http://jdmlm.ub.ac.id/index.php/jdmlm/oai, pertama di-harvest 9 Nov 2026, terakhir terlihat 16 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">Assessing the impact of training samples overlap and density in random forest for landslide susceptibility mapping: Implications for degraded land management in Bandung Regency, Indonesia</dc:title>
	<dc:creator>Nugroho, Hary</dc:creator>
	<dc:creator>Sari, Dewi Kania</dc:creator>
	<dc:creator>Safitri, Sitarani</dc:creator>
	<dc:creator>Azmi, Naufal</dc:creator>
	<dc:subject xml:lang="en-US">hyperparameter tuning</dc:subject>
	<dc:subject xml:lang="en-US">landslide susceptibility  mapping </dc:subject>
	<dc:subject xml:lang="en-US">random forest</dc:subject>
	<dc:subject xml:lang="en-US">overlapping training sample</dc:subject>
	<dc:description xml:lang="en-US">Landslide susceptibility mapping is essential for disaster mitigation and land management in degraded mountainous regions. Machine learning algorithms, particularly Random Forest (RF), have been increasingly applied due to their robustness in handling complex, non-linear relationships. However, classification performance is often affected by the quality of training samples, especially when landslide and non-landslide points exhibit spatial overlap. This study investigated how varying densities of fully overlapping samples influence RF performance in Bandung Regency, West Java, Indonesia, an area characterised by steep slopes, rapid land-use change, and post-mining degradation. Balanced datasets ranging from 50 to 700 samples per class were evaluated with hyperparameter tuning. The highest validation accuracy (89%) was achieved with 500 samples at a max_depth of 2, while training accuracy was approximately 10% lower, indicating the algorithm’s difficulty in separating overlapping classes. A more stable trade-off was obtained with 300 samples and a max_depth of 4, suggesting that moderate densities enhance generalisation. To translate these findings into practice, we propose an ensemble zoning and uncertainty mapping framework that integrates multiple model outputs to identify consensus zones for slope stabilisation, vegetation restoration, and adaptive spatial planning. This approach improves the reliability of susceptibility maps and provides actionable insights for managing degraded and landslide-prone landscapes.</dc:description>
	<dc:publisher xml:lang="en-US">Brawijaya University</dc:publisher>
	<dc:date>2025-10-01</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://jdmlm.ub.ac.id/index.php/jdmlm/article/view/17311</dc:identifier>
	<dc:identifier>10.15243/jdmlm.2025.125.8933</dc:identifier>
	<dc:source xml:lang="en-US">Journal of Degraded and Mining Lands Management; Vol. 12 No. 5 (2025); 8933-8956</dc:source>
	<dc:source>2502-2458</dc:source>
	<dc:source>2339-076X</dc:source>
	<dc:source>10.15243/jdmlm.2025.125</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://jdmlm.ub.ac.id/index.php/jdmlm/article/view/17311/1732</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2025 Journal of Degraded and Mining Lands Management</dc:rights>
	<dc:rights xml:lang="en-US">https://creativecommons.org/licenses/by-nc/4.0</dc:rights>
</oai_dc:dc>