Next Article in Journal
Copyright Protection Based on Zero Watermarking and Blockchain for Vector Maps
Previous Article in Journal
Analysis of the Degree of Threat to Railway Infrastructure by Falling Tree Vegetation
Article

Assessing the Generalization of Machine Learning-Based Slope Failure Prediction to New Geographic Extents

1
Department of Geology and Geography, West Virginia University, Morgantown, WV 26505, USA
2
West Virginia GIS Technical Center, Morgantown, WV 26505, USA
*
Author to whom correspondence should be addressed.
Academic Editors: Biswajeet Pradhan and Wolfgang Kainz
ISPRS Int. J. Geo-Inf. 2021, 10(5), 293; https://doi.org/10.3390/ijgi10050293
Received: 17 March 2021 / Revised: 18 April 2021 / Accepted: 1 May 2021 / Published: 3 May 2021
Slope failure probabilistic models generated using random forest (RF) machine learning (ML), manually interpreted incident points, and light detection and ranging (LiDAR) digital terrain variables are assessed for predicting and generalizing to new geographic extents. Specifically, models for four Major Land Resource Areas (MLRAs) in the state of West Virginia in the United States (US) were created. All region-specific models were then used to predict withheld validation data within all four MLRAs. For all validation datasets, the model trained using data from the same MLRA provided the highest reported overall accuracy (OA), Kappa statistic, F1 Score, area under the receiver operating characteristic curve (AUC ROC), and area under the precision-recall curve (AUC PR). However, the model from the same MLRA as the validation dataset did not always provide the highest precision, recall, and/or specificity, suggesting that models extrapolated to new geographic extents tend to either overpredict or underpredict the land area of slope failure occurrence whereas they offer a better balance between omission and commission error within the region in which they were trained. This study highlights the value of developing region-specific inventories, models, and high resolution and detailed digital elevation data, since models may not generalize well to new geographic extents, potentially resulting from spatial heterogeneity in landscape and/or slope failure characteristics. View Full-Text
Keywords: slope failures; landslides; light detection and ranging; LiDAR; digital terrain analysis; machine learning; random forest; spatial predictive modeling; generalization slope failures; landslides; light detection and ranging; LiDAR; digital terrain analysis; machine learning; random forest; spatial predictive modeling; generalization
Show Figures

Figure 1

MDPI and ACS Style

Maxwell, A.E.; Sharma, M.; Kite, J.S.; Donaldson, K.A.; Maynard, S.M.; Malay, C.M. Assessing the Generalization of Machine Learning-Based Slope Failure Prediction to New Geographic Extents. ISPRS Int. J. Geo-Inf. 2021, 10, 293. https://doi.org/10.3390/ijgi10050293

AMA Style

Maxwell AE, Sharma M, Kite JS, Donaldson KA, Maynard SM, Malay CM. Assessing the Generalization of Machine Learning-Based Slope Failure Prediction to New Geographic Extents. ISPRS International Journal of Geo-Information. 2021; 10(5):293. https://doi.org/10.3390/ijgi10050293

Chicago/Turabian Style

Maxwell, Aaron E., Maneesh Sharma, J. S. Kite, Kurt A. Donaldson, Shannon M. Maynard, and Caleb M. Malay 2021. "Assessing the Generalization of Machine Learning-Based Slope Failure Prediction to New Geographic Extents" ISPRS International Journal of Geo-Information 10, no. 5: 293. https://doi.org/10.3390/ijgi10050293

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop