Perbandingan Model Machine Learning untuk Prediksi Bencana Tanah Longsor Secara Spasial di Kabupaten Malang
Kata Kunci:
tanah longsor, logistic regression, random forest, SVM, analisis distribusi spasial, analisis korelasiAbstrak
Penelitian ini bertujuan untuk mengembangkan model prediksi kerawanan tanah longsor di Kabupaten Malang, Jawa Timur, dengan memanfaatkan model machine learning berbasis data geospasial dan metodologi CRISP-ML(Q). Rumusan masalah yang diangkat mencakup: (1) bagaimana hasil perbandingan akurasi prediksi tanah longsor berdasarkan model Support Vector Machine (SVM), Random Forest (RF), dan Logistic Regression (LR); (2) bagaimana sebaran spasial hasil prediksi tanah longsor di Kabupaten Malang berdasarkan masing-masing model; dan (3) apa saja faktor yang memengaruhi kerawanan suatu desa terhadap tanah longsor. Hasil analisis menunjukkan bahwa model SVM memiliki akurasi tertinggi sebesar 0,99, diikuti oleh LR dengan akurasi 0,97, dan RF yang mencapai akurasi 0,86. Pemetaan kerawanan tanah longsor mengidentifikasi daerah di sekitar Gunung Bromo dan utara Gunung Kawi sebagai wilayah paling rentan, dengan model SVM mengklasifikasikan 36,22% desa sebagai daerah rawan. Analisis korelasi menunjukkan bahwa data historis longsor, tutupan pepohonan, kerapatan vegetasi, dan kemiringan lereng berkorelasi positif terhadap kerawanan, sedangkan jenis geologi dan rata-rata curah hujan berkorelasi negatif. Temuan ini diharapkan dapat menjadi pertimbangan bagi pemerintah daerah dalam merumuskan strategi mitigasi bencana yang lebih efektif dan meningkatkan kesiapsiagaan masyarakat terhadap risiko tanah longsor.
Referensi
Agboola, G., Beni, L.H., Elbayoumi, T. and Thompson, G., 2024. Optimizing landslide susceptibility mapping using machine learning and geospatial techniques. Ecological Informatics, 81, p.102583. https://doi.org/10.1016/J.ECOINF.2024.102583.
Badan Penanggulangan Bencana Daerah, 2023. Banyaknya Bencana Alam Menurut Jumlah Korban dan Kerugian di Kabupaten Malang: Kamasuta Bertali. [online] Kamasuta Bertali: Kabupaten Malang Satu Data Berbasis Teknologi Informasi. Available at: <http://kamasuta.malangkab.go.id/data-nilai-public?id=7517> [Accessed 28 August 2024].
Lyu, H.M., Yin, Z.Y., Hicher, P.Y. and Laouafa, F., 2024. Incorporating mitigation strategies in machine learning for landslide susceptibility prediction. Geoscience Frontiers, 15(5), p.101869. https://doi.org/10.1016/J.GSF.2024.101869.
Md.Sharafat, C., Md.Naimur, R., Md.Sujon, S., Md.Abu, S., Mahmud, K.H. and Hafsa, B., 2024. GIS-based landslide susceptibility mapping using logistic regression, random forest and decision and regression tree models in Chattogram District, Bangladesh. Heliyon, 10(1), p.e23424. https://doi.org/10.1016/J.HELIYON.2023.E23424.
Mohri, M., Rostamizadeh, A. and Talwalkar, A., 2018. Foundations of Machine Learning second edition. 2nd ed. Cambridge: The MIT Press.
PDSI Pusdatinkom BNPB, 2024. Data Bencana Indonesia 2023. [online] Jakarta. Available at: <https://bnpb.go.id/buku/buku-data-bencana-indonesia-tahun-2023> [Accessed 28 August 2024].
Prathom, K. and Sujitapan, C., 2024. Performance of logistic regression and support vector machine conjunction with the GIS and RS in the landslide susceptibility assessment: Case study in Nakhon Si Thammarat, southern Thailand. Journal of King Saud University - Science, 36(8), p.103306. https://doi.org/10.1016/J.JKSUS.2024.103306.
Putri, N.H., Dananjaya, R.H. and Surjandari, N.S., 2024. Aplikasi Machine Learning Method pada Pemetaan Kerawanan Tanah Longsor di Kabupaten Karanganyar. Media Komunikasi Dunia Ilmu Sipil (MoDuluS), [online] 6(1), pp.23–34. https://doi.org/10.32585/MODULUS.V6I1.4490.
Ramdani, F., 2017. Pengantar Ilmu Geoinformatika. Malang: UB Press.
Retnowati, Rr.E., 2021. PEMETAAN RISIKO BENCANA TANAH LONGSOR DAN PREDIKSI KERUGIAN PETANI BADAN PERTANAHAN NASIONAL REPUBLIK INDONESIA SEKOLAH TINGGI PERTANAHAN NASIONAL YOGYAKARTA 2014. [online] Sekolah Tinggi Pertanahan Nasional. Available at: <https://repository.stpn.ac.id/1313/1/Rr.%20Endah%20Retnowati.pdf> [Accessed 4 September 2024].
Setiawan, E.B., 2020. Sistem Informasi Geografis Berbasis Web Menggunakan Google Maps dan Mapbox API. Bandung: Informatika Bandung.
Sharma, A. and Singh Sandhu, H.A., 2023. GIS-Based Landslide Susceptibility Analysis Using Machine Learning Algorithms. In: IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium. IEEE. pp.3038–3041. https://doi.org/10.1109/IGARSS52108.2023.10283099.
Sinaga, B.A., Vionanda, D., Permana, D. and Salma, D.A., 2023. Comparison of Error Rate Prediction Methods in Binary Logistic Regression Modeling for Imbalanced Data. UNP Journal of Statistics and Data Science, [online] 1(4), pp.361–368. https://doi.org/10.24036/UJSDS/VOL1-ISS4/86.
Studer, S., Bui, T.B., Drescher, C., Hanuschkin, A., Winkler, L., Peters, S. and Müller, K.R., 2021. Towards CRISP-ML(Q): A Machine Learning Process Model with Quality Assurance Methodology. Machine Learning and Knowledge Extraction 2021, Vol. 3, Pages 392-413, [online] 3(2), pp.392–413. https://doi.org/10.3390/MAKE3020020.
Suyanto, 2018. Machine Learning Tingkat Dasar dan Lanjut. Bandung: Informatika Bandung.
Watt, J., Borhani, R. and Katsaggelos, A.K., 2020. Machine Learning Refined Foundations, Algorithms, and Applications (Jeremy Watt Reza Borhani Aggelos Katsaggelos). 2nd ed. Cambridge: Cambridge University Press.
Wu, W., Guo, S. and Shao, Z., 2023. Landslide risk evaluation and its causative factors in typical mountain environment of China: a case study of Yunfu City. Ecological Indicators, 154, p.110821. https://doi.org/10.1016/J.ECOLIND.2023.110821.
Zhang, A., Zhao, X. wen, Zhao, X. yuezi, Zheng, X. zhan, Zeng, M., Huang, X., Wu, P., Jiang, T., Wang, S. chang, He, J. and Li, Y. yong, 2024. Comparative study of different machine learning models in landslide susceptibility assessment: A case study of Conghua District, Guangzhou, China. China Geology, 7(1), pp.104–115. https://doi.org/10.31035/CG2023056.
Agboola, G., Beni, L.H., Elbayoumi, T. and Thompson, G., 2024. Optimizing landslide susceptibility mapping using machine learning and geospatial techniques. Ecological Informatics, 81, p.102583. https://doi.org/10.1016/J.ECOINF.2024.102583.
Badan Penanggulangan Bencana Daerah, 2023. Banyaknya Bencana Alam Menurut Jumlah Korban dan Kerugian di Kabupaten Malang: Kamasuta Bertali. [online] Kamasuta Bertali: Kabupaten Malang Satu Data Berbasis Teknologi Informasi. Available at: <http://kamasuta.malangkab.go.id/data-nilai-public?id=7517> [Accessed 28 August 2024].
Lyu, H.M., Yin, Z.Y., Hicher, P.Y. and Laouafa, F., 2024. Incorporating mitigation strategies in machine learning for landslide susceptibility prediction. Geoscience Frontiers, 15(5), p.101869. https://doi.org/10.1016/J.GSF.2024.101869.
Md.Sharafat, C., Md.Naimur, R., Md.Sujon, S., Md.Abu, S., Mahmud, K.H. and Hafsa, B., 2024. GIS-based landslide susceptibility mapping using logistic regression, random forest and decision and regression tree models in Chattogram District, Bangladesh. Heliyon, 10(1), p.e23424. https://doi.org/10.1016/J.HELIYON.2023.E23424.
Mohri, M., Rostamizadeh, A. and Talwalkar, A., 2018. Foundations of Machine Learning second edition. 2nd ed. Cambridge: The MIT Press.
PDSI Pusdatinkom BNPB, 2024. Data Bencana Indonesia 2023. [online] Jakarta. Available at: <https://bnpb.go.id/buku/buku-data-bencana-indonesia-tahun-2023> [Accessed 28 August 2024].
Prathom, K. and Sujitapan, C., 2024. Performance of logistic regression and support vector machine conjunction with the GIS and RS in the landslide susceptibility assessment: Case study in Nakhon Si Thammarat, southern Thailand. Journal of King Saud University - Science, 36(8), p.103306. https://doi.org/10.1016/J.JKSUS.2024.103306.
Putri, N.H., Dananjaya, R.H. and Surjandari, N.S., 2024. Aplikasi Machine Learning Method pada Pemetaan Kerawanan Tanah Longsor di Kabupaten Karanganyar. Media Komunikasi Dunia Ilmu Sipil (MoDuluS), [online] 6(1), pp.23–34. https://doi.org/10.32585/MODULUS.V6I1.4490.
Ramdani, F., 2017. Pengantar Ilmu Geoinformatika. Malang: UB Press.
Retnowati, Rr.E., 2021. PEMETAAN RISIKO BENCANA TANAH LONGSOR DAN PREDIKSI KERUGIAN PETANI BADAN PERTANAHAN NASIONAL REPUBLIK INDONESIA SEKOLAH TINGGI PERTANAHAN NASIONAL YOGYAKARTA 2014. [online] Sekolah Tinggi Pertanahan Nasional. Available at: <https://repository.stpn.ac.id/1313/1/Rr.%20Endah%20Retnowati.pdf> [Accessed 4 September 2024].
Setiawan, E.B., 2020. Sistem Informasi Geografis Berbasis Web Menggunakan Google Maps dan Mapbox API. Bandung: Informatika Bandung.
Sharma, A. and Singh Sandhu, H.A., 2023. GIS-Based Landslide Susceptibility Analysis Using Machine Learning Algorithms. In: IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium. IEEE. pp.3038–3041. https://doi.org/10.1109/IGARSS52108.2023.10283099.
Sinaga, B.A., Vionanda, D., Permana, D. and Salma, D.A., 2023. Comparison of Error Rate Prediction Methods in Binary Logistic Regression Modeling for Imbalanced Data. UNP Journal of Statistics and Data Science, [online] 1(4), pp.361–368. https://doi.org/10.24036/UJSDS/VOL1-ISS4/86.
Studer, S., Bui, T.B., Drescher, C., Hanuschkin, A., Winkler, L., Peters, S. and Müller, K.R., 2021. Towards CRISP-ML(Q): A Machine Learning Process Model with Quality Assurance Methodology. Machine Learning and Knowledge Extraction 2021, Vol. 3, Pages 392-413, [online] 3(2), pp.392–413. https://doi.org/10.3390/MAKE3020020.
Suyanto, 2018. Machine Learning Tingkat Dasar dan Lanjut. Bandung: Informatika Bandung.
Watt, J., Borhani, R. and Katsaggelos, A.K., 2020. Machine Learning Refined Foundations, Algorithms, and Applications (Jeremy Watt Reza Borhani Aggelos Katsaggelos). 2nd ed. Cambridge: Cambridge University Press.
Wu, W., Guo, S. and Shao, Z., 2023. Landslide risk evaluation and its causative factors in typical mountain environment of China: a case study of Yunfu City. Ecological Indicators, 154, p.110821. https://doi.org/10.1016/J.ECOLIND.2023.110821.
Zhang, A., Zhao, X. wen, Zhao, X. yuezi, Zheng, X. zhan, Zeng, M., Huang, X., Wu, P., Jiang, T., Wang, S. chang, He, J. and Li, Y. yong, 2024. Comparative study of different machine learning models in landslide susceptibility assessment: A case study of Conghua District, Guangzhou, China. China Geology, 7(1), pp.104–115. https://doi.org/10.31035/CG2023056.
Unduhan
Diterbitkan
Cara Mengutip
Terbitan
Bagian
Lisensi
Hak Cipta (c) 2025 Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Artikel ini berlisensiCreative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
