“…Over the last two decades, thanks to increasingly successful analyses of the relationship between past landslide events and environmental characteristics, statistical and machine learning models have been increasingly widely applied to predict landslide susceptibility. Statistical models include weight of evidence (Achu et al, 2022; Hong et al, 2017; Lee & Choi, 2004), logistic regression (Budimir et al, 2015; Lee, 2005), frequency ratio (Achu et al, 2023; Lee & Pradhan, 2007; Li et al, 2017; Pradhan, 2010a) and fuzzy logic (Gopinath et al, 2023; Pourghasemi et al, 2012; Pradhan, 2011). Although these models have been applied extensively in previous studies to predict landslide susceptibility, landslides have a complex, nonlinear nature and are dependent on a variety of factors, for example, environment, climate, hydrology and human activity: there is a limit to the accuracy achievable using statistical models in place.…”
Landslides lead to widespread devastation and significant loss of life in mountainous regions around the world. Susceptibility assessments can provide critical data to help decision‐makers, for example, local authorities and other organizations, mitigating the landslide risk, although the accuracy of existing studies needs to be improved. This study aims to assess landslide susceptibility in the Thua Thien Hue province of Vietnam using deep neural networks (DNNs) and swarm‐based optimization algorithms, namely Adam, stochastic gradient descent (SGD), Artificial Rabbits Optimization (ARO), Tuna Swarm Optimization (TSO), Sand Cat Swarm Optimization (SCSO), Honey Badger Algorithm (HBA), Marine Predators Algorithm (MPA) and Particle Swarm Optimization (PSO). The locations of 945 landslides occurring between 2012 and 2022, along with 14 conditioning factors, were used as input data to build the DNN and DNN‐hybrid models. The performance of the proposed models was evaluated using the statistical indices receiver operating characteristic curve, area under the curve (AUC), root mean square error, mean absolute error (MAE), R2 and accuracy. All proposed models had a high accuracy of prediction. The DNN‐MPA model had the highest AUC value (0.95), followed by DNN‐HBA (0.95), DNN‐ARO (0.95), DNN‐Adam (0.95), DNN‐SGD (0.95), DNN‐TSO (0.93), DNN‐PSO (0.9) and finally DNN‐SCSO (0.83). High‐precision models have identified that the majority of the western region of Thua Thien Hue province is very highly susceptible to landslides. Models like the aforementioned ones can support decision‐makers in updating large‐scale sustainable land‐use strategies.
“…Over the last two decades, thanks to increasingly successful analyses of the relationship between past landslide events and environmental characteristics, statistical and machine learning models have been increasingly widely applied to predict landslide susceptibility. Statistical models include weight of evidence (Achu et al, 2022; Hong et al, 2017; Lee & Choi, 2004), logistic regression (Budimir et al, 2015; Lee, 2005), frequency ratio (Achu et al, 2023; Lee & Pradhan, 2007; Li et al, 2017; Pradhan, 2010a) and fuzzy logic (Gopinath et al, 2023; Pourghasemi et al, 2012; Pradhan, 2011). Although these models have been applied extensively in previous studies to predict landslide susceptibility, landslides have a complex, nonlinear nature and are dependent on a variety of factors, for example, environment, climate, hydrology and human activity: there is a limit to the accuracy achievable using statistical models in place.…”
Landslides lead to widespread devastation and significant loss of life in mountainous regions around the world. Susceptibility assessments can provide critical data to help decision‐makers, for example, local authorities and other organizations, mitigating the landslide risk, although the accuracy of existing studies needs to be improved. This study aims to assess landslide susceptibility in the Thua Thien Hue province of Vietnam using deep neural networks (DNNs) and swarm‐based optimization algorithms, namely Adam, stochastic gradient descent (SGD), Artificial Rabbits Optimization (ARO), Tuna Swarm Optimization (TSO), Sand Cat Swarm Optimization (SCSO), Honey Badger Algorithm (HBA), Marine Predators Algorithm (MPA) and Particle Swarm Optimization (PSO). The locations of 945 landslides occurring between 2012 and 2022, along with 14 conditioning factors, were used as input data to build the DNN and DNN‐hybrid models. The performance of the proposed models was evaluated using the statistical indices receiver operating characteristic curve, area under the curve (AUC), root mean square error, mean absolute error (MAE), R2 and accuracy. All proposed models had a high accuracy of prediction. The DNN‐MPA model had the highest AUC value (0.95), followed by DNN‐HBA (0.95), DNN‐ARO (0.95), DNN‐Adam (0.95), DNN‐SGD (0.95), DNN‐TSO (0.93), DNN‐PSO (0.9) and finally DNN‐SCSO (0.83). High‐precision models have identified that the majority of the western region of Thua Thien Hue province is very highly susceptible to landslides. Models like the aforementioned ones can support decision‐makers in updating large‐scale sustainable land‐use strategies.
Landslides are among the most destructive geological disasters that seriously damage human life and infrastructures. Landslides mostly occur in mountainous regions around the world. One of the key processes to reduce these damages is to uncover landslide-exposed areas through different data-driven methods such as Geographical Information System (GIS) and Multi-Criteria Decision-Making (MCDM). In the literature, there are many studies developed with these fundamental tools. In this study, unlike the literature, a new landslide susceptibility assessment model is proposed by integrating GIS with the stratified best-worst method (S-BWM). This model has four main dimensions and 16 sub-dimensions under topography, environment-land, location and hydrological factors, which are weighted with the S-BWM. Considering the different states that may arise in the importance weights of these dimensions in the future, a network was created. The transition probabilities of these states were predicted and injected into the classical BWM. Then maps were created for these dimensions and classifications were made for each subdimension according to the map characteristics. Finally, the most susceptive landslide locations were determined with GIS-based calculations. To demonstrate the model's applicability, a case study was conducted for the Erzurum region, one of Turkey's landslide-prone regions. In addition, besides the landslide map, an analysis and discussion about the spatial distribution of susceptibility classes was presented, contributing to the study's robustness.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.