Spatial prediction models for landslide hazards: review, comparison and evaluation

By Alexander Brenning. 2005

654 citations recorded by the source corpus.

Abstract. The predictive power of logistic regression, support vector machines and bootstrap-aggregated classification trees (bagging, double-bagging) is compared using misclassification error rates on independent test data sets. Based on a resampling approach that takes into account spatial autocorrelation, error rates for predicting "present" and "future" landslides are estimated within and outside the training area. In a case study from the Ecuadorian Andes, logistic regression with stepwise backward variable selection yields lowest error rates and demonstrates the best generalization capabilities. The evaluation outside the training area reveals that tree-based methods tend to overfit the data.

Subfields: Atmospheric Science, Environmental Engineering, Management, Monitoring, Policy and Law

View Paper: https://www.semanticscholar.org/paper/d04a9a9e72cbf162660316b7b4ae8fc8832304c7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

In Environmental Engineering

In Management, Monitoring, Policy and Law

Citations in the corpus

Landslide susceptibility mapping based on GIS and support vector machine models for the Qianyang County, China (2016, methodology): “Brenning (2005) obtained sufficiently smooth prediction surfaces for creating susceptibility map by using SVM. Yao et al. (2008) used the SVM in landslide susceptibility mapping, they found that SVM was a useful tool in landslide susceptibility assessment, and they found that the SVM had better…”
Ecology of butterflies in Europe (2009, background): “A number of studies comparing different modelling techniques could not identify any single method to be superior for all species and situations (Guisan et al., 1999; Manel et al., 1999; Moisen & Frescino, 2002; Olden & Jackson, 2002; Segurado & Araújo, 2004; Brenning, 2005; Moisen et al., 2006; Pott”
Landslide susceptibility mapping at Hoa Binh province (Vietnam) using an adaptive neuro-fuzzy inference system and GIS (2012, neutral): “(Brenning,2005; ChungandFabbri,1999,2003;Leeetal.,2007).”
Reasons for an outstanding plant diversity in the tropical Andes of southern Ecuador. (2009, background): “The geologic structure and petrographic situation of the RBSF-terrain giving rise to the formation of steep slopes promote an ongoing landslide activity, which is obvious on 8,5 % of the area shown in Figure 13 (Brenning 2005, based on data from Stoyan 2000).”
How does forest structure affect root reinforcement and susceptibility to shallow landslides? (2016, positive): “Brenning (2005) and Schicker and Moon (2012) also found better generalization capabilities for logistic regression compared to other classification models.”
Importance of spatial predictor variable selection in machine learning applications - Moving from data reproduction to spatial prediction (2019, methodology): “Several methods for spatial cross-validation have been proposed to account for spatial dependencies in the data (Brenning, 2005; Le Rest et al., 2014; Pohjankukka et al., 2017; Roberts et al., 2017; Meyer et al., 2018; Valavi et al., 2018).”
Accounting for covariate distributions in slope-unit-based landslide susceptibility models. A case study in the alpine environment (2019, methodology): “…data for the susceptibility model.257 4 Statistical modeling258 Spatial predictive models represent the primary tool for assessing landslide susceptibility at259 catchment or regional scales (Brenning, 2005), although the choice of the best mapping unit260 is not always straightforward.”
Landslide susceptibility evaluation and hazard zonation techniques – a review (2020, methodology): “Support vector machines (SVM) This method is used to solve complex classification and regression problems (Tien Bui et al. 2015; Kavzoglu et al. 2013; Ballabio and Sterlacchini 2012; Kavzoglu and Colkesen 2009; Brenning 2005).”
An easy-to-use MATLAB program (MamLand) for the assessment of landslide susceptibility using a Mamdani fuzzy algorithm (2012, methodology): “Computers & Geosciences (2011) by varying the decision threshold (Brenning, 2005) In the AUC method, the AUC, with values ranging from 0.5 to 1.0, is used to assess the accuracy of the constructed model.”
Spatial prediction of groundwater spring potential mapping based on an adaptive neuro-fuzzy inference system and metaheuristic optimization (2018, methodology): “Therefore, it can be used for model prediction capabilities (Brenning, 2005).”