Abstract
Landslides pose a constant threat to the lives and property of mountain people and may also cause geomorphological destruction such as soil and water loss, vegetation destruction, and land cover change. Landslide susceptibility assessment (LSA) is a key component of landslide risk evaluation. There are many related studies, but few analyses and comparisons of models for optimization. This paper aims to introduce the Tree-structured Parzen Estimator (TPE) algorithm for hyperparameter optimization of three typical neural network models for LSA in Shuicheng County, China, as an example, and to compare the differences of predictive ability among the models in order to achieve higher application performance. First, 17 influencing factors of landslide multiple data sources were selected for spatial prediction, hybrid ensemble oversampling and undersampling techniques were used to address the imbalanced sample and small sample size problem, and the samples were randomly divided into a training set and validation set. Second, deep neural network (DNN), recurrent neural network (RNN), and convolutional neural network (CNN) models were adopted to predict the regional landslides susceptibility, and the TPE algorithm was used to optimize the hyperparameters respectively to improve the assessment capacity. Finally, to compare the differences and optimization effects of these models, several objective measures were applied for validation. The results show that the high-susceptibility regions mostly distributed in bands along fault zones, where the lithology is mostly claystone, sandstone, and basalt. The DNN, RNN, and CNN models all perform well in LSA, especially the RNN model. The TPE optimization significantly improves the accuracy of the DNN and CNN (3.92% and 1.52%, respectively), but does not improve the performance of the RNN. In summary, our proposed RNN model and TPE-optimized DNN and CNN model have robust predictive capability for landslide susceptibility in the study area and can also be applied to other areas containing similar geological conditions.
Highlights
Topple is the crumbling and rolling of a rock and soil body on a slope after it has been suddenly detached by gravity, slide is the overall downward sliding of a rock body on a slope under the action of gravity for some reason along a certain weak surface or zone of weakness, debris flow is a special type of flood with large quantities of sediment, rocks and other solid material conditions formed by precipitation
The results indicate that no multicollinearity relationship existed among factors
Before aims to provide an introduction for application and comparison of three typical neural training of these models, for the preparation of the sample set, due to the insufficient network models (DNN, recurrent neural network (RNN), convolutional neural network (CNN)) in Landslide susceptibility assessment (LSA), and to optimize their hyperparameters using number of positive samples, this study proposed to use the hybrid ensemble oversamTPE algorithm in order to get better prediction accuracy and performance
Summary
China is a country where landslides occur very frequently and the damage is extremely serious. There are about one million historical landslide sites, including topple, slide, debris flow, ground subsidence, and other types. Topples, slides, and debris flows constitute 80% of the whole landslides [3]. The trigger conditions and thresholds of the three landslide types are different, the geological and hydrological conditions (susceptibility factors) of the areas where they may occur are extremely similar, and they are more frequent and hazardous. Topples, slides, and debris flows are integrated to represent landslide for susceptibility assessment in this paper
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
