A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Assessing landslide susceptibility based on hybrid multilayer perceptron with ensemble learning
https://orcid.org/0000-0001-6224-069X
Abstract Landslides have brought about serious human and economic losses worldwide. Modeling landslide susceptibility is an important technology to avoid the loss caused by landslide disasters. The aim of this study was to design three integrated models by combining the multilayer perceptron (MLP) model with three meta classifiers (Decorate, MultiboostAB, and Rotation Forest) for modeling landslide susceptibility. Yanshan County was selected as the case study. The multicollinearity diagnoses Spearman’s and ReliefF were applied to judge the proper factors for modeling landslide susceptibility, and the results demonstrate that plan curvature and topographic wetness index are not proper factors in Yanshan County; hence, these factors were not used. Based on several statistical indices, the three integrated models achieve better results than the MLP model. MultiboostAB-MLP is a more stable and effective model, and the three integrated models were better than the existing machine learning algorithm. The result demonstrates that the overall trend of change is stable using different sampling ratios between the training and validation data. Finally, the landslide susceptibility map produced by the three integrated models may be applied for land use management and building reconstruction.
Assessing landslide susceptibility based on hybrid multilayer perceptron with ensemble learning
https://orcid.org/0000-0001-6224-069X
Abstract Landslides have brought about serious human and economic losses worldwide. Modeling landslide susceptibility is an important technology to avoid the loss caused by landslide disasters. The aim of this study was to design three integrated models by combining the multilayer perceptron (MLP) model with three meta classifiers (Decorate, MultiboostAB, and Rotation Forest) for modeling landslide susceptibility. Yanshan County was selected as the case study. The multicollinearity diagnoses Spearman’s and ReliefF were applied to judge the proper factors for modeling landslide susceptibility, and the results demonstrate that plan curvature and topographic wetness index are not proper factors in Yanshan County; hence, these factors were not used. Based on several statistical indices, the three integrated models achieve better results than the MLP model. MultiboostAB-MLP is a more stable and effective model, and the three integrated models were better than the existing machine learning algorithm. The result demonstrates that the overall trend of change is stable using different sampling ratios between the training and validation data. Finally, the landslide susceptibility map produced by the three integrated models may be applied for land use management and building reconstruction.
Assessing landslide susceptibility based on hybrid multilayer perceptron with ensemble learning
https://orcid.org/0000-0001-6224-069X
Abstract Landslides have brought about serious human and economic losses worldwide. Modeling landslide susceptibility is an important technology to avoid the loss caused by landslide disasters. The aim of this study was to design three integrated models by combining the multilayer perceptron (MLP) model with three meta classifiers (Decorate, MultiboostAB, and Rotation Forest) for modeling landslide susceptibility. Yanshan County was selected as the case study. The multicollinearity diagnoses Spearman’s and ReliefF were applied to judge the proper factors for modeling landslide susceptibility, and the results demonstrate that plan curvature and topographic wetness index are not proper factors in Yanshan County; hence, these factors were not used. Based on several statistical indices, the three integrated models achieve better results than the MLP model. MultiboostAB-MLP is a more stable and effective model, and the three integrated models were better than the existing machine learning algorithm. The result demonstrates that the overall trend of change is stable using different sampling ratios between the training and validation data. Finally, the landslide susceptibility map produced by the three integrated models may be applied for land use management and building reconstruction.
Assessing landslide susceptibility based on hybrid multilayer perceptron with ensemble learning
Hong, Haoyuan (author)
2023
Article (Journal)
Electronic Resource
English
BKL:
56.00$jBauwesen: Allgemeines
/
38.58
Geomechanik
/
38.58$jGeomechanik
/
56.20
Ingenieurgeologie, Bodenmechanik
/
56.00
Bauwesen: Allgemeines
/
56.20$jIngenieurgeologie$jBodenmechanik
RVK:
ELIB18
Improved Shallow Landslide Susceptibility Prediction Based on Statistics and Ensemble Learning
| DOAJ | 2022
Landslide Susceptibility Assessment by Novel Hybrid Machine Learning Algorithms
| DOAJ | 2019
Application of novel ensemble models to improve landslide susceptibility mapping reliability
| Online Contents | 2023