Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment
To address the errors of negative samples in landslide susceptibility modeling and traditional methods in exploring the regularities hidden in the evaluation factors, this paper proposes a stacking one- and three-dimensional Convolutional Neural Network (Stacking-1D-3D-CNN) landslide susceptibility...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-01-01
|
| Series: | Geo-spatial Information Science |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/10095020.2024.2443483 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841525416008876032 |
|---|---|
| author | Yueyue Wang Xueling Wu Kun Zhou Guo Lin Bo Peng Zhice Fang |
| author_facet | Yueyue Wang Xueling Wu Kun Zhou Guo Lin Bo Peng Zhice Fang |
| author_sort | Yueyue Wang |
| collection | DOAJ |
| description | To address the errors of negative samples in landslide susceptibility modeling and traditional methods in exploring the regularities hidden in the evaluation factors, this paper proposes a stacking one- and three-dimensional Convolutional Neural Network (Stacking-1D-3D-CNN) landslide susceptibility assessment method considering sample optimization selection. First, in order to select negative samples rationally, this paper adopts the Relative Frequency Ratio combined with Certainty Factor Method (RFR-CFM) to determine the negative samples; secondly, the Stacking-1D-3D-CNN proposed is combined with RFR-CFM for the first time for landslide susceptibility assessment. In this work, the negative samples determined by RFR-CFM and the Information Quality Model (IQM) were combined with historical disaster points to form a total modeling sample, and modeled at different ratios. Finally, it is compared with several other models in terms of the landslide hazard susceptibility zoning results, prone zone statistics, and model performance. The findings show that the degree of spatial aggregation of training samples and testing samples has a much greater impact on the accuracy of landslide susceptibility modeling than the impact of their proportions. Furthermore, compared with other models, RFR-CFM-Stacking-1D-3D-CNN has the highest AUC value, precision, recall, F-score, and accuracy, which are 0.95, 0.83, 0.89, 0.85, and 84.76%, respectively, and the lowest RMSE and MAE, 0.39 and 0.15, respectively. This proves the RFR-CFM sample selection method’s rationality and the Stacking-1D-3D-CNN model’s effectiveness. |
| format | Article |
| id | doaj-art-dc8f8d170ef242ee8dd38a2186bb0ff0 |
| institution | Kabale University |
| issn | 1009-5020 1993-5153 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Geo-spatial Information Science |
| spelling | doaj-art-dc8f8d170ef242ee8dd38a2186bb0ff02025-01-17T14:21:56ZengTaylor & Francis GroupGeo-spatial Information Science1009-50201993-51532025-01-0112110.1080/10095020.2024.2443483Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessmentYueyue Wang0Xueling Wu1Kun Zhou2Guo Lin3Bo Peng4Zhice Fang5School of Geophysics and Geomatics, China University of Geosciences, Wuhan, ChinaSchool of Geophysics and Geomatics, China University of Geosciences, Wuhan, ChinaSchool of Electrical and Information Engineering, Hubei University of Automotive Technology, Shiyan, ChinaDepartment of Atmospheric and Oceanic Science, University of Colorado Boulder, Boulder, USASchool of Geophysics and Geomatics, China University of Geosciences, Wuhan, ChinaSchool of Geophysics and Geomatics, China University of Geosciences, Wuhan, ChinaTo address the errors of negative samples in landslide susceptibility modeling and traditional methods in exploring the regularities hidden in the evaluation factors, this paper proposes a stacking one- and three-dimensional Convolutional Neural Network (Stacking-1D-3D-CNN) landslide susceptibility assessment method considering sample optimization selection. First, in order to select negative samples rationally, this paper adopts the Relative Frequency Ratio combined with Certainty Factor Method (RFR-CFM) to determine the negative samples; secondly, the Stacking-1D-3D-CNN proposed is combined with RFR-CFM for the first time for landslide susceptibility assessment. In this work, the negative samples determined by RFR-CFM and the Information Quality Model (IQM) were combined with historical disaster points to form a total modeling sample, and modeled at different ratios. Finally, it is compared with several other models in terms of the landslide hazard susceptibility zoning results, prone zone statistics, and model performance. The findings show that the degree of spatial aggregation of training samples and testing samples has a much greater impact on the accuracy of landslide susceptibility modeling than the impact of their proportions. Furthermore, compared with other models, RFR-CFM-Stacking-1D-3D-CNN has the highest AUC value, precision, recall, F-score, and accuracy, which are 0.95, 0.83, 0.89, 0.85, and 84.76%, respectively, and the lowest RMSE and MAE, 0.39 and 0.15, respectively. This proves the RFR-CFM sample selection method’s rationality and the Stacking-1D-3D-CNN model’s effectiveness.https://www.tandfonline.com/doi/10.1080/10095020.2024.2443483Landslide susceptibilityrelative frequency ratio (RFR)certainty factor method (CFM)stackingconvolutional neural network (CNN) |
| spellingShingle | Yueyue Wang Xueling Wu Kun Zhou Guo Lin Bo Peng Zhice Fang Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment Geo-spatial Information Science Landslide susceptibility relative frequency ratio (RFR) certainty factor method (CFM) stacking convolutional neural network (CNN) |
| title | Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment |
| title_full | Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment |
| title_fullStr | Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment |
| title_full_unstemmed | Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment |
| title_short | Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment |
| title_sort | integrating a multi dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment |
| topic | Landslide susceptibility relative frequency ratio (RFR) certainty factor method (CFM) stacking convolutional neural network (CNN) |
| url | https://www.tandfonline.com/doi/10.1080/10095020.2024.2443483 |
| work_keys_str_mv | AT yueyuewang integratingamultidimensionaldeepconvolutionalneuralnetworkwithoptimizedsampleselectionforlandslidesusceptibilityassessment AT xuelingwu integratingamultidimensionaldeepconvolutionalneuralnetworkwithoptimizedsampleselectionforlandslidesusceptibilityassessment AT kunzhou integratingamultidimensionaldeepconvolutionalneuralnetworkwithoptimizedsampleselectionforlandslidesusceptibilityassessment AT guolin integratingamultidimensionaldeepconvolutionalneuralnetworkwithoptimizedsampleselectionforlandslidesusceptibilityassessment AT bopeng integratingamultidimensionaldeepconvolutionalneuralnetworkwithoptimizedsampleselectionforlandslidesusceptibilityassessment AT zhicefang integratingamultidimensionaldeepconvolutionalneuralnetworkwithoptimizedsampleselectionforlandslidesusceptibilityassessment |