Comparison of landslide susceptibility maps using random forest and multivariate adaptive regression spline models in combination with catchment map units

2018 ◽  
Vol 23 (2) ◽  
pp. 341-355 ◽  
Author(s):  
Lei Chu ◽  
Liang-Jie Wang ◽  
Jiang Jiang ◽  
Xia Liu ◽  
Kazuhide Sawada ◽  
...  
Keyword(s):  
Random Forest ◽  
Landslide Susceptibility ◽  
Multivariate Adaptive Regression Spline ◽  
Regression Spline ◽  
Spline Models ◽  
Susceptibility Maps ◽  
Adaptive Regression ◽  
Landslide Susceptibility Maps

Landslide Susceptibility Mapping in the Commune of Oudka, Taounate Province, North Morocco: A Comparative Analysis of Logistic Regression, Multivariate Adaptive Regression Spline, and Artificial Neural Network Models

2020 ◽  
Vol 26 (2) ◽  
pp. 185-200
Author(s):  
Said Benchelha ◽  
Hasnaa Chennaoui Aoudjehane ◽  
Mustapha Hakdaoui ◽  
Rachid El Hamdouni ◽  
Hamou Mansouri ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Landslide Susceptibility ◽  
Vegetation Index ◽  
Multivariate Adaptive Regression Spline ◽  
Landslide Susceptibility Mapping ◽  
Slope Aspect ◽  
Neural Network Models ◽  
Regression Spline ◽  
Adaptive Regression ◽  
Artificial Neural

ABSTRACT Landslide susceptibility indices were calculated and landslide susceptibility maps were generated for the Oudka, Morocco, study area using a geographic information system. The spatial database included current landslide location, topography, soil, hydrology, and lithology, and the eight factors related to landslides (elevation, slope, aspect, distance to streams, distance to roads, distance to faults, lithology, and Normalized Difference Vegetation Index [NDVI]) were calculated or extracted. Logistic regression (LR), multivariate adaptive regression spline (MARSpline), and Artificial Neural Networks (ANN) were the methods used in this study to generate landslide susceptibility indices. Before the calculation, the study area was randomly divided into two parts, the first for the establishment of the model and the second for its validation. The results of the landslide susceptibility analysis were verified using success and prediction rates. The MARSpline model gave a higher success rate (AUC (Area Under The Curve) = 0.963) and prediction rate (AUC = 0.951) than the LR model (AUC = 0.918 and AUC = 0.901) and the ANN model (AUC = 0.886 and AUC = 0.877). These results indicate that the MARSpline model is the best model for determining landslide susceptibility in the study area.

Coupling Multivariate Adaptive Regression Spline (MARS) and Random Forest (RF)

2019 ◽  
Vol 14 (1) ◽  
pp. 1-18 ◽  
Author(s):  
Arpita Nagpal ◽  
Vijendra Singh
Keyword(s):  
Feature Selection ◽  
Random Forest ◽  
Classification Accuracy ◽  
Hybrid Approach ◽  
Multivariate Adaptive Regression Spline ◽  
Comparative Performance ◽  
Maximum Information ◽  
Regression Spline ◽  
Adaptive Regression ◽  
Selection Algorithms

In this article, a new algorithm to select the relevant features is proposed for handling microarray data with the specific aim of increasing classification accuracy. In particular, the optimal genes are extracted using filter and wrapper feature selection algorithms. Here, the use of non-parametric regression algorithm called Multivariate Adaptive Regression Spline (MARS) followed by proposed Random Forest Statistical Test (RFST) algorithm are being studied. The study evaluates the comparative performance of the results of RFST and MARS with existing algorithms on ten standard microarray datasets. For performance analysis, three parameters are taken into consideration, namely, the number of selected features, runtime, and classification accuracy. Experimental results indicate that different feature selection algorithms yield different candidate gene subset; therefore, a Hybrid approach is applied to determine the best candidate genes to provide maximum information about the disease. The findings foretell that the RFST is performing better on six out of ten datasets whereas MARS is performing better on other datasets.

scholarly journals A Novel Performance Assessment Approach using Photogrammetric Techniques for Landslide Susceptibility Mapping with Logistic Regression, ANN and Random Forest

Sensors ◽  
2019 ◽  
Vol 19 (18) ◽  
pp. 3940 ◽  
Author(s):  
Sevgen ◽  
Kocaman ◽  
Nefeslioglu ◽  
Gokceoglu
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Random Forest ◽  
Performance Assessment ◽  
Landslide Susceptibility ◽  
Machine Learning Algorithms ◽  
Landslide Susceptibility Mapping ◽  
Susceptibility Maps ◽  
Surface Models ◽  
Landslide Susceptibility Maps

Prediction of possible landslide areas is the first stage of landslide hazard mitigation efforts and is also crucial for suitable site selection. Several statistical and machine learning methodologies have been applied for the production of landslide susceptibility maps. However, the performance assessment of such methods have conventionally been carried out by utilizing existing landslide inventories. The purpose of this study is to investigate the performances of landslide susceptibility maps produced with three different machine learning algorithms, i.e., random forest, artificial neural network, and logistic regression, in a recently constructed and activated dam reservoir and assess the external quality of each map by using pre- and post-event photogrammetric datasets. The methodology introduced here was applied using digital surface models generated from aerial photogrammetric flight data acquired before and after the dam construction. Aerial photogrammetric images acquired in 2012 and 2018 (after the dam was filled) were used to produce digital terrain models and orthophotos. The 2012 dataset was used for producing the landslide susceptibility maps and the results were evaluated by comparing the Euclidian distances between the two surface models. The results show that the random forest method outperforms the other two for predicting the future landslides.

Sign in / Sign up

Export Citation Format

Share Document