Landslide susceptibility modelling using GIS-based machine learning techniques for Chongren County, Jiangxi Province, China

Chen, Wei; Peng, Jianbing; Hong, Haoyuan; Shahabi, Himan; Pradhan, Biswajeet; Liu, Junzhi; Zhu, A‐Xing; Pei, Xiangjun; Duan, Zhao

doi:10.1016/j.scitotenv.2018.01.124

Cited by 346 publications

(200 citation statements)

References 46 publications

Supporting

Mentioning

166

Contrasting

Unclassified

Order By: Relevance

“…For poor prediction or lack of improvement, the area under the curve (AUC) values become less or equal to 0.5, whereas the ideal susceptibility modeling is obtained when AUC values are equal to or greater than 0.7 (Chen et al., ). The AUROC curves were confirmed as one of the most common and useful gold standard tools to validate and compare landslide susceptibility modeling methods in recent studies (Chen et al., ; Zêzere et al., ). To assess and compare the predictive capabilities of the two models, the statistical measures were calculated using Equations – (Chen et al., ; Tharwat, in press):

AUC = \frac{((), \sum TP + \sum TN)}{(P + N)},

Accuracy = \frac{TP + TN}{TP + FP + TN + FN},

Precision = \frac{TP}{TP + FP},

where P is the total number of landslides, N is the total number of nonlandslides, TP (true positive) and TN (true negative) are the numbers of pixels that are classified correctly, and FP (false positive) and FN (false negative) are the numbers of pixels classified incorrectly (Chen et al., ).…”

Section: Methodsmentioning

confidence: 88%

“…Therefore, the ROC presents the percentages of true positive ratings of past landslides against the false-positive rating percentage of the susceptibility index in a cumulative, decreasing order. This result is used to obtain the ROC curves of the rate of success (Ahmed & Dewan, 2017;Chen et al, 2018a). The area under the ROC curve (AUROC) is therefore used to detect model performance in predicting landslide susceptibility for the study area.…”

Section: Model Performance Evaluation and Validationmentioning

confidence: 99%

“…Landslides are among the deadliest natural disasters throughout the world (Chen et al, 2018a;Chen, Zhang, Li, & Shahabi, 2018b). Thus, the losses and fatalities induced by landslide hazards are numerous.…”

Section: Introductionmentioning

confidence: 99%

“…The literature on landslide studies confirms that different models used to predict landslide susceptibility have been widely compared globally (Chen et al, 2018a;Juliev et al, 2018). Qualitative models and data-driven models-namely, statistical methodswere explored and compared to study landslide susceptibility for different regional scales (Dou et al, 2018;Van Den Eeckhaut, Marre, & Poesen, 2010).…”

Section: Introductionmentioning

confidence: 99%

“…The weights are assigned to conditioning factors based on landslide density. In addition, in recent comparative studies on landslide susceptibility modeling, statistical models were explored and compared with machine-learning techniques (Chen et al, 2018a;Goetz, Brenning, Petschko, & Leopold, 2015). Generally, statistical methods showed better performance for landslide susceptibility prediction worldwide.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Comparative Analysis of Deterministic and Semiquantitative Approaches for Shallow Landslide Risk Modeling in Rwanda

et al. 2019

View full text Add to dashboard Cite

The use of appropriate approaches to produce risk maps is critical in landslide disaster management. The aim of this study was to investigate and compare the stability index mapping (SINMAP) and the spatial multicriteria evaluation (SMCE) models for landslide risk modeling in Rwanda. The SINMAP used the digital elevation model in conjunction with physical soil parameters to determine the factor of safety. The SMCE method used six layers of landslide conditioning factors. In total, 155 past landslide locations were used for training and model validation. The results showed that the SMCE performed better than the SINMAP model. Thus, the receiver operating characteristic and three statistical estimators—accuracy, precision, and the root mean square error (RMSE)—were used to validate and compare the predictive capabilities of the two models. Therefore, the area under the curve (AUC) values were 0.883 and 0.798, respectively, for the SMCE and SINMAP. In addition, the SMCE model produced the highest accuracy and precision values of 0.770 and 0.734, respectively. For the RMSE values, the SMCE produced better prediction than SINMAP (0.332 and 0.398, respectively). The overall comparison of results confirmed that both SINMAP and SMCE models are promising approaches for landslide risk prediction in central‐east Africa.

show abstract

AUC = \frac{((), \sum TP + \sum TN)}{(P + N)},

Accuracy = \frac{TP + TN}{TP + FP + TN + FN},

Precision = \frac{TP}{TP + FP},

Section: Methodsmentioning

confidence: 88%

Section: Model Performance Evaluation and Validationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Comparative Analysis of Deterministic and Semiquantitative Approaches for Shallow Landslide Risk Modeling in Rwanda

et al. 2019

View full text Add to dashboard Cite

show abstract

Insights into landslide susceptibility in different karst erosion landforms based on interpretable machine learning

Sun,

Wang,

Wen

et al. 2023

Earth Surf Processes Landf

View full text Add to dashboard Cite

The aim of the present study was to assess differences in the conditioning factors and the performance of landslide susceptibility mapping (LSM), employing the SHapley Additive exPlanations (SHAP) model to gain profound insights into the intrinsic decision‐making mechanism of LSM in diverse landforms. Two typical karst erosion landforms were selected as the research areas. Based on 15 conditioning factors, LSMs for the two areas were developed using the Bayesian optimization random forest (RF) and eXtreme Gradient Boosting (XGBoost). The SHAP model was used to explore the landslide formation mechanisms from both global and local perspectives. The results show that the area under the curve (AUC) values of the XGBoost models were 0.791 and 0.761, and the AUC values of the RF models were 0.844 and 0.817, in the two different landform areas, respectively. The RF model's accuracy was higher than that of the XGBoost model in both regions. In the low‐elevation hills area, the primary three conditioning factors were identified as slope, topographic relief and distance from the river. Conversely, in the microrelief and mesorelief low mountain area, the predominant conditioning factors were elevation, distance from the river and distance from the road. Both karst landform areas exhibited a high sensitivity to the distance from the river, indicating its significant interaction with other factors contributing to landslide occurrences. Notably, the RF model demonstrated superior performance compared to the XGBoost model, rendering it a more suitable choice for conducting landslide susceptibility mapping research in karst erosion landform areas. In the present study, a comprehensive explanatory framework based on the RF‐SHAP model was proposed, which enables both global and local interpretation of landslides in various karst landscapes. Such an approach explores the intrinsic decision‐making mechanism of the model, enhancing the transparency and realism of landslide susceptibility prediction results.

show abstract

Landslide susceptibility prediction using machine learning and remote sensing: Case study in Thua Thien Hue province, Vietnam

Nguyen,

et al. 2023

Geological Journal

View full text Add to dashboard Cite

Landslides lead to widespread devastation and significant loss of life in mountainous regions around the world. Susceptibility assessments can provide critical data to help decision‐makers, for example, local authorities and other organizations, mitigating the landslide risk, although the accuracy of existing studies needs to be improved. This study aims to assess landslide susceptibility in the Thua Thien Hue province of Vietnam using deep neural networks (DNNs) and swarm‐based optimization algorithms, namely Adam, stochastic gradient descent (SGD), Artificial Rabbits Optimization (ARO), Tuna Swarm Optimization (TSO), Sand Cat Swarm Optimization (SCSO), Honey Badger Algorithm (HBA), Marine Predators Algorithm (MPA) and Particle Swarm Optimization (PSO). The locations of 945 landslides occurring between 2012 and 2022, along with 14 conditioning factors, were used as input data to build the DNN and DNN‐hybrid models. The performance of the proposed models was evaluated using the statistical indices receiver operating characteristic curve, area under the curve (AUC), root mean square error, mean absolute error (MAE), R2 and accuracy. All proposed models had a high accuracy of prediction. The DNN‐MPA model had the highest AUC value (0.95), followed by DNN‐HBA (0.95), DNN‐ARO (0.95), DNN‐Adam (0.95), DNN‐SGD (0.95), DNN‐TSO (0.93), DNN‐PSO (0.9) and finally DNN‐SCSO (0.83). High‐precision models have identified that the majority of the western region of Thua Thien Hue province is very highly susceptible to landslides. Models like the aforementioned ones can support decision‐makers in updating large‐scale sustainable land‐use strategies.

show abstract

Landslide susceptibility modelling using GIS-based machine learning techniques for Chongren County, Jiangxi Province, China

Cited by 346 publications

References 46 publications

Comparative Analysis of Deterministic and Semiquantitative Approaches for Shallow Landslide Risk Modeling in Rwanda

Comparative Analysis of Deterministic and Semiquantitative Approaches for Shallow Landslide Risk Modeling in Rwanda

Insights into landslide susceptibility in different karst erosion landforms based on interpretable machine learning

Landslide susceptibility prediction using machine learning and remote sensing: Case study in Thua Thien Hue province, Vietnam

Contact Info

Product

Resources

About