Impact of orthogonal transformation for factors on model performance in landslide susceptibility

Considering the correlation among the conditioning factors, this study explores the impact of the orthogonal transformation for factors on model performance, with the LR model. Landslide inventory and factor theme maps were first constructed and used for factor analysis. Subsequently, effective fact...

Full description

Saved in:
Bibliographic Details
Published in:Environmental earth sciences 2023-03, Vol.82 (5), p.119, Article 119
Main Authors: Liu, Qiang, Tang, Aiping, Huang, Delong, Sun, Lixin
Format: Article
Language:English
Subjects:
Accuracy
Biogeosciences
Civil engineering
Coefficients
Computer applications
Construction
Correlation
Earth and Environmental Science
Earth science
Earth Sciences
Environmental Science and Engineering
Factor analysis
Geochemistry
Geology
Hydrology/Water Resources
Landslides
Landslides & mudslides
Modelling
Optimization
Original Article
Performance prediction
Plate tectonics
Regression coefficients
Sensitivity
Specificity
Terrestrial Pollution
Topography
Transformations
Workflow
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Considering the correlation among the conditioning factors, this study explores the impact of the orthogonal transformation for factors on model performance, with the LR model. Landslide inventory and factor theme maps were first constructed and used for factor analysis. Subsequently, effective factors with high correlation were applied for orthogonal transformation, to obtain the reconstructed factors. Meanwhile, reconstructed factors and the remaining effective factors were jointly involved in the model construction. Finally, differences in model performance before and after the transformation were discussed, in terms of the predictive ability and computational ability. Results indicate that the modified model maintains high predictive ability compared with the initial model, i.e., high values of AUC (0.939), specificity (0.956), sensitivity (0.977), and accuracy (0.968) in the training phase, and AUC (0.928), specificity (0.791), sensitivity (0.866), and accuracy (0.849) in the test phase. Besides, the modified model not only exhibits high operation speed but also smaller standard errors of the regression coefficients compared with the initial model, overall. The high predictive and computational performance of the modified susceptibility model provides a window for optimization of the landslide susceptibility in the modeling workflow.
ISSN:1866-6280
1866-6299
DOI:10.1007/s12665-023-10803-8