Machine Learning Feature Selection Methods for Landslide Susceptibility Mapping
- PDF / 1,890,528 Bytes
- 25 Pages / 439.37 x 666.142 pts Page_size
- 5 Downloads / 241 Views
Machine Learning Feature Selection Methods for Landslide Susceptibility Mapping Natan Micheletti · Loris Foresti · Sylvain Robert · Michael Leuenberger · Andrea Pedrazzini · Michel Jaboyedoff · Mikhail Kanevski
Received: 4 March 2013 / Accepted: 23 November 2013 © International Association for Mathematical Geosciences 2013
Abstract This paper explores the use of adaptive support vector machines, random forests and AdaBoost for landslide susceptibility mapping in three separated regions of Canton Vaud, Switzerland, based on a set of geological, hydrological and morphological features. The feature selection properties of the three algorithms are studied to analyze the relevance of features in controlling the spatial distribution of landslides. The elimination of irrelevant features gives simpler, lower dimensional models while keeping the classification performance high. An object-based sampling procedure is considered to reduce the spatial autocorrelation of data and to estimate more reliably generalization skills when applying the model to predict the occurrence of new unknown landslides. The accuracy of the models, the relevance of features and the quality of landslide susceptibility maps were found to be high in the regions characterized by shallow landslides and low in the ones with deep-seated landslides. Despite providing similar skill, random forests and AdaBoost were found to be more efficient in performing feature selection than adaptive support vector machines. The results of this study reveal the strengths of the classification algorithms, but evidence: (1) the need for relying on more than one method for the identification of relevant variables;
N. Micheletti (B) · M. Leuenberger · M. Kanevski Institute of Earth Surface Dynamics, Faculty of Geosciences and Environment, University of Lausanne, 1015 Lausanne, Switzerland e-mail: [email protected] L. Foresti Royal Meteorological Institute of Belgium, 1180 Brussels, Belgium S. Robert Seminar for Statistics, Swiss Federal Institute of Technology, 8092 Zürich, Switzerland A. Pedrazzini · M. Jaboyedoff Institute of Earth Sciences, Faculty of Geosciences and Environment, University of Lausanne, 1015 Lausanne, Switzerland
123
Math Geosci
(2) the weakness of the adaptive scaling algorithm when used with landslide data; and (3) the lack of additional features which characterize the spatial distribution of deep-seated landslides. Keywords Landslide susceptibility mapping · Support vector machines (SVM) · Adaptive scaling SVM · Random forests · AdaBoost · Multiscale terrain features · Object-based validation
1 Introduction Territorial planning and natural hazards assessment make extensive use of landslide susceptibility (LS) maps. Geotechnical, physical and statistical approaches can be designed to predict the occurrence of slope failures (Montgomery and Dietrich 1994; Soeters and van Westen 1996; van Westen et al. 2005, 2008). However, geotechnical methods are more suited to study specific events and are often applied to single slopes (Daia and Lee 20
Data Loading...