Application of Bayesian hyperparameter optimized random forest and XGBoost model for landslide susceptibility mapping
TLDR
In this article, the authors used Bayesian hyperparameters to optimize random forest and extreme gradient boosting decision trees model for landslide susceptibility mapping, and the two optimized models are compared using the receiver operating characteristic curve and confusion matrix.Abstract:
Landslides are widely distributed worldwide and often result in tremendous casualties and economic losses, especially in the Loess Plateau of China. Taking Wuqi County in the hinterland of the Loess Plateau as the research area, using Bayesian hyperparameters to optimize random forest and extreme gradient boosting decision trees model for landslide susceptibility mapping, and the two optimized models are compared. In addition, 14 landslide influencing factors are selected, and 734 landslides are obtained according to field investigation and reports from literals. The landslides were randomly divided into training data (70%) and validation data (30%). The hyperparameters of the random forest and extreme gradient boosting decision tree models were optimized using a Bayesian algorithm, and then the optimal hyperparameters are selected for landslide susceptibility mapping. Both models were evaluated and compared using the receiver operating characteristic curve and confusion matrix. The results show that the AUC validation data of the Bayesian optimized random forest and extreme gradient boosting decision tree model are 0.88 and 0.86, respectively, which showed an improvement of 4% and 3%, indicating that the prediction performance of the two models has been improved. However, the random forest model has a higher predictive ability than the extreme gradient boosting decision tree model. Thus, hyperparameter optimization is of great significance in the improvement of the prediction accuracy of the model. Therefore, the optimized model can generate a high-quality landslide susceptibility map.read more
Citations
More filters
Journal ArticleDOI
Landslide Susceptibility mapping using random forest and extreme gradient boosting: A case study of Fengjie, Chongqing
TL;DR: In this paper , the evaluation effects of random forest (RF) and extreme gradient boosting (XGBoost) classifier models on landslide susceptibility, and to compare their applicability in Fengjie County, Chongqing, a typical landslideprone area in southwest of China.
Journal ArticleDOI
Application of Tree-Based Ensemble Models to Landslide Susceptibility Mapping: A Comparative Study
TL;DR: The results presented in this study indicate that the advanced ensemble model, the XGBoost model, could be a promising tool for the selection of ensemble models for predicting landslide susceptibility mapping.
Journal ArticleDOI
Advanced hyperparameter optimization for improved spatial prediction of shallow landslides using extreme gradient boosting (XGBoost)
Taskin Kavzoglu,Alihan Teke +1 more
TL;DR: Analysis of computational cost efficiency and AUC analysis showed that the Hyperband approach was much faster than the GA in hyperparameter tuning, and thus appeared to be the best optimization algorithm for the problem under consideration.
Journal ArticleDOI
Landslide Susceptibility Assessment Model Construction Using Typical Machine Learning for the Three Gorges Reservoir Area in China
TL;DR: Wang et al. as mentioned in this paper applied machine learning models, including logistic regression (LR), the random forest model (RF), and the support vector machine (SVM) model, to assess landslide susceptibility in the Yangtze River's Three Gorges Reservoir region to analyze landslide events in the whole study region.
Journal ArticleDOI
Evaluation of landslide susceptibility of the Ya’an–Linzhi section of the Sichuan–Tibet Railway based on deep learning
TL;DR: Wang et al. as mentioned in this paper used two deep learning algorithms, CNN and DNN, to map the landslide susceptibility of the Ya'an-Lin branch of the Sichuan-Tibet Railway.
References
More filters
Journal ArticleDOI
Random Forests
TL;DR: Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the forest, and are also applicable to regression.
Journal ArticleDOI
Guidelines for landslide susceptibility, hazard and risk zoning for land-use planning
TL;DR: In this article, the authors presented a study of the relationship between geotechnical engineering and geosciences and geophysics at the University of New South Wales and U.S. Geological Survey.
Journal ArticleDOI
A review of statistically-based landslide susceptibility models
TL;DR: In this paper, a critical review of statistical methods for landslide susceptibility modelling and associated terrain zonations is presented, revealing a significant heterogeneity of thematic data types and scales, modelling approaches, and model evaluation criteria.
Journal ArticleDOI
GIS techniques and statistical models in evaluating landslide hazard
TL;DR: In this article, a small drainage basin located in Central Italy, relevant geological and geomorphological factors were collected and processed by applying GIS technology, which both generate high-fidelity digital terrain models and automatically partition the terrain into main slope-units.
Journal ArticleDOI
Global fatal landslide occurrence from 2004 to 2016
Abstract: . Landslides are a ubiquitous hazard in terrestrial environments with slopes,
incurring human fatalities in urban settlements, along transport corridors
and at sites of rural industry. Assessment of landslide risk requires
high-quality landslide databases. Recently, global landslide databases have
shown the extent to which landslides impact on society and identified areas
most at risk. Previous global analysis has focused on rainfall-triggered
landslides over short ∼ 5-year observation periods. This paper presents
spatiotemporal analysis of a global dataset of fatal non-seismic landslides,
covering the period from January 2004 to December 2016. The data show that in
total 55 997 people were killed in
4862 distinct landslide events. The spatial distribution of landslides
is heterogeneous, with Asia representing the dominant geographical area.
There are high levels of interannual variation in the occurrence of
landslides. Although more active years coincide with recognised patterns of
regional rainfall driven by climate anomalies, climate modes (such as El
Nino–Southern Oscillation) cannot yet be related to landsliding,
requiring a landslide dataset of 30 + years. Our analysis demonstrates that
landslide occurrence triggered by human activity is increasing, in particular
in relation to construction, illegal mining and hill cutting. This supports
notions that human disturbance may be more detrimental to future landslide
incidence than climate.