Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping

Abstract Deterministic models have been widely applied in landslide risk assessment (LRA), but they have limitations in obtaining various geotechnical and hydraulic properties. The objective of this study is to suggest a new deterministic method based on machine learning (ML) algorithms. Eight cruci...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Dae-Hong Min, Hyung-Koo Yoon
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/23486316c6a74f17ba1620bd2de21987
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:23486316c6a74f17ba1620bd2de21987
record_format dspace
spelling oai:doaj.org-article:23486316c6a74f17ba1620bd2de219872021-12-02T16:36:04ZSuggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping10.1038/s41598-021-86137-x2045-2322https://doaj.org/article/23486316c6a74f17ba1620bd2de219872021-03-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-86137-xhttps://doaj.org/toc/2045-2322Abstract Deterministic models have been widely applied in landslide risk assessment (LRA), but they have limitations in obtaining various geotechnical and hydraulic properties. The objective of this study is to suggest a new deterministic method based on machine learning (ML) algorithms. Eight crucial variables of LRA are selected with reference to expert opinions, and the output value is set to the safety factor derived by Mohr–Coulomb failure theory in infinite slope. Linear regression and a neural network based on ML are applied to find the best model between independent and dependent variables. To increase the reliability of linear regression and the neural network, the results of back propagation, including gradient descent, Levenberg–Marquardt (LM), and Bayesian regularization (BR) methods, are compared. An 1800-item dataset is constructed through measured data and artificial data by using a geostatistical technique, which can provide the information of an unknown area based on measured data. The results of linear regression and the neural network show that the special LM and BR back propagation methods demonstrate a high determination of coefficient. The important variables are also investigated though random forest (RF) to overcome the number of various input variables. Only four variables—shear strength, soil thickness, elastic modulus, and fine content—demonstrate a high reliability for LRA. The results show that it is possible to perform LRA with ML, and four variables are enough when it is difficult to obtain various variables.Dae-Hong MinHyung-Koo YoonNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-24 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Dae-Hong Min
Hyung-Koo Yoon
Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping
description Abstract Deterministic models have been widely applied in landslide risk assessment (LRA), but they have limitations in obtaining various geotechnical and hydraulic properties. The objective of this study is to suggest a new deterministic method based on machine learning (ML) algorithms. Eight crucial variables of LRA are selected with reference to expert opinions, and the output value is set to the safety factor derived by Mohr–Coulomb failure theory in infinite slope. Linear regression and a neural network based on ML are applied to find the best model between independent and dependent variables. To increase the reliability of linear regression and the neural network, the results of back propagation, including gradient descent, Levenberg–Marquardt (LM), and Bayesian regularization (BR) methods, are compared. An 1800-item dataset is constructed through measured data and artificial data by using a geostatistical technique, which can provide the information of an unknown area based on measured data. The results of linear regression and the neural network show that the special LM and BR back propagation methods demonstrate a high determination of coefficient. The important variables are also investigated though random forest (RF) to overcome the number of various input variables. Only four variables—shear strength, soil thickness, elastic modulus, and fine content—demonstrate a high reliability for LRA. The results show that it is possible to perform LRA with ML, and four variables are enough when it is difficult to obtain various variables.
format article
author Dae-Hong Min
Hyung-Koo Yoon
author_facet Dae-Hong Min
Hyung-Koo Yoon
author_sort Dae-Hong Min
title Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping
title_short Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping
title_full Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping
title_fullStr Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping
title_full_unstemmed Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping
title_sort suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/23486316c6a74f17ba1620bd2de21987
work_keys_str_mv AT daehongmin suggestionforanewdeterministicmodelcoupledwithmachinelearningtechniquesforlandslidesusceptibilitymapping
AT hyungkooyoon suggestionforanewdeterministicmodelcoupledwithmachinelearningtechniquesforlandslidesusceptibilitymapping
_version_ 1718383619403153408