scholarly journals A Hybrid Model Consisting of Supervised and Unsupervised Learning for Landslide Susceptibility Mapping

2021 ◽  
Vol 13 (8) ◽  
pp. 1464
Author(s):  
Zhu Liang ◽  
Changming Wang ◽  
Zhijie Duan ◽  
Hailiang Liu ◽  
Xiaoyang Liu ◽  
...  
Keyword(s):  
Unsupervised Learning ◽  
Hybrid Model ◽  
Landslide Susceptibility ◽  
Learning Model ◽  
Susceptibility Mapping ◽  
Landslide Susceptibility Mapping ◽  
Gradient Boosting ◽  
Susceptibility Map ◽  
Continuous Variables ◽  
Human Beings

Landslides cause huge damage to social economy and human beings every year. Landslide susceptibility mapping (LSM) occupies an important position in land use and risk management. This study is to investigate a hybrid model which makes full use of the advantage of supervised learning model (SLM) and unsupervised learning model (ULM). Firstly, ten continuous variables were used to develop a ULM which consisted of factor analysis (FA) and k-means cluster for a preliminary landslide susceptibility map. Secondly, 351 landslides with “1” label were collected and the same number of non-landslide samples with “0” label were selected from the very low susceptibility area in the preliminary map, constituting a new priori condition for a SLM, and thirteen factors were used for the modeling of gradient boosting decision tree (GBDT) which represented for SLM. Finally, the performance of different models was verified using related indexes. The results showed that the performance of the pretreated GBDT model was improved with sensitivity, specificity, accuracy and the area under the curve (AUC) values of 88.60%, 92.59%, 90.60% and 0.976, respectively. It can be concluded that a pretreated model with strong robustness can be constructed by increasing the purity of samples.

scholarly journals A Novel Hybrid Method for Landslide Susceptibility Mapping-Based GeoDetector and Machine Learning Cluster: A Case of Xiaojin County, China

2021 ◽  
Vol 10 (2) ◽  
pp. 93
Author(s):  
Wei Xie ◽  
Xiaoshuang Li ◽  
Wenbin Jian ◽  
Yang Yang ◽  
Hongwei Liu ◽  
...  
Keyword(s):  
Machine Learning ◽  
Hybrid Method ◽  
Roc Curve ◽  
Landslide Susceptibility ◽  
Susceptibility Mapping ◽  
Assessment Model ◽  
Landslide Susceptibility Mapping ◽  
Support Vector ◽  
Susceptibility Map ◽  
Area Index

Landslide susceptibility mapping (LSM) could be an effective way to prevent landslide hazards and mitigate losses. The choice of conditional factors is crucial to the results of LSM, and the selection of models also plays an important role. In this study, a hybrid method including GeoDetector and machine learning cluster was developed to provide a new perspective on how to address these two issues. We defined redundant factors by quantitatively analyzing the single impact and interactive impact of the factors, which was analyzed by GeoDetector, the effect of this step was examined using mean absolute error (MAE). The machine learning cluster contains four models (artificial neural network (ANN), Bayesian network (BN), logistic regression (LR), and support vector machines (SVM)) and automatically selects the best one for generating LSM. The receiver operating characteristic (ROC) curve, prediction accuracy, and the seed cell area index (SCAI) methods were used to evaluate these methods. The results show that the SVM model had the best performance in the machine learning cluster with the area under the ROC curve of 0.928 and with an accuracy of 83.86%. Therefore, SVM was chosen as the assessment model to map the landslide susceptibility of the study area. The landslide susceptibility map demonstrated fit with landslide inventory, indicated the hybrid method is effective in screening landslide influences and assessing landslide susceptibility.

scholarly journals Incorporating Landslide Spatial Information and Correlated Features among Conditioning Factors for Landslide Susceptibility Mapping

2021 ◽  
Vol 13 (11) ◽  
pp. 2166
Author(s):  
Xin Yang ◽  
Rui Liu ◽  
Mei Yang ◽  
Jingjue Chen ◽  
Tianqiang Liu ◽  
...  
Keyword(s):  
Land Use ◽  
Hybrid Model ◽  
Landslide Susceptibility ◽  
Spatial Information ◽  
Landslide Susceptibility Mapping ◽  
Ratio Method ◽  
Support Vector ◽  
Susceptibility Map ◽  
Conditioning Factors ◽  
Proposed Model

This study proposed a new hybrid model based on the convolutional neural network (CNN) for making effective use of historical datasets and producing a reliable landslide susceptibility map. The proposed model consists of two parts; one is the extraction of landslide spatial information using two-dimensional CNN and pixel windows, and the other is to capture the correlated features among the conditioning factors using one-dimensional convolutional operations. To evaluate the validity of the proposed model, two pure CNN models and the previously used methods of random forest and a support vector machine were selected as the benchmark models. A total of 621 earthquake-triggered landslides in Ludian County, China and 14 conditioning factors derived from the topography, geological, hydrological, geophysical, land use and land cover data were used to generate a geospatial dataset. The conditioning factors were then selected and analyzed by a multicollinearity analysis and the frequency ratio method. Finally, the trained model calculated the landslide probability of each pixel in the study area and produced the resultant susceptibility map. The results indicated that the hybrid model benefitted from the features extraction capability of the CNN and achieved high-performance results in terms of the area under the receiver operating characteristic curve (AUC) and statistical indices. Moreover, the proposed model had 6.2% and 3.7% more improvement than the two pure CNN models in terms of the AUC, respectively. Therefore, the proposed model is capable of accurately mapping landslide susceptibility and providing a promising method for hazard mitigation and land use planning. Additionally, it is recommended to be applied to other areas of the world.

scholarly journals Comparison between Deep Learning and Tree-Based Machine Learning Approaches for Landslide Susceptibility Mapping

Water ◽  
2021 ◽  
Vol 13 (19) ◽  
pp. 2664
Author(s):  
Sunil Saha ◽  
Jagabandhu Roy ◽  
Tusar Kanti Hembram ◽  
Biswajeet Pradhan ◽  
Abhirup Dikshit ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Deep Learning ◽  
Landslide Susceptibility ◽  
Learning Model ◽  
Susceptibility Mapping ◽  
Landslide Susceptibility Mapping ◽  
Learning Approaches ◽  
Statistical Measures ◽  
Deep Learning Model

The efficiency of deep learning and tree-based machine learning approaches has gained immense popularity in various fields. One deep learning model viz. convolution neural network (CNN), artificial neural network (ANN) and four tree-based machine learning models, namely, alternative decision tree (ADTree), classification and regression tree (CART), functional tree and logistic model tree (LMT), were used for landslide susceptibility mapping in the East Sikkim Himalaya region of India, and the results were compared. Landslide areas were delimited and mapped as landslide inventory (LIM) after gathering information from historical records and periodic field investigations. In LIM, 91 landslides were plotted and classified into training (64 landslides) and testing (27 landslides) subsets randomly to train and validate the models. A total of 21 landslide conditioning factors (LCFs) were considered as model inputs, and the results of each model were categorised under five susceptibility classes. The receiver operating characteristics curve and 21 statistical measures were used to evaluate and prioritise the models. The CNN deep learning model achieved the priority rank 1 with area under the curve of 0.918 and 0.933 by using the training and testing data, quantifying 23.02% and 14.40% area as very high and highly susceptible followed by ANN, ADtree, CART, FTree and LMT models. This research might be useful in landslide studies, especially in locations with comparable geophysical and climatological characteristics, to aid in decision making for land use planning.

scholarly journals Landslide susceptibility mapping by using a geographic information system (GIS) along the China–Pakistan Economic Corridor (Karakoram Highway), Pakistan

2019 ◽  
Vol 19 (5) ◽  
pp. 999-1022 ◽  
Author(s):  
Sajid Ali ◽  
Peter Biermanns ◽  
Rashid Haider ◽  
Klaus Reicherter
Keyword(s):  
Information System ◽  
Geographic Information System ◽  
Landslide Susceptibility ◽  
Slope Angle ◽  
Active Faults ◽  
Geographic Information ◽  
Susceptibility Mapping ◽  
Landslide Susceptibility Mapping ◽  
Susceptibility Map ◽  
Karakoram Highway

Abstract. The Karakoram Highway (KKH) is an important route, which connects northern Pakistan with Western China. Presence of steep slopes, active faults and seismic zones, sheared rock mass, and torrential rainfall make the study area a unique geohazards laboratory. Since its construction, landslides constitute an appreciable threat, having blocked the KKH several times. Therefore, landslide susceptibility mapping was carried out in this study to support highway authorities in maintaining smooth and hazard-free travelling. Geological and geomorphological data were collected and processed using a geographic information system (GIS) environment. Different conditioning and triggering factors for landslide occurrences were considered for preparation of the susceptibility map. These factors include lithology, seismicity, rainfall intensity, faults, elevation, slope angle, aspect, curvature, land cover and hydrology. According to spatial and statistical analyses, active faults, seismicity and slope angle mainly control the spatial distribution of landslides. Each controlling parameter was assigned a numerical weight by utilizing the analytic hierarchy process (AHP) method. Additionally, the weighted overlay method (WOL) was employed to determine landslide susceptibility indices. As a result, the landslide susceptibility map was produced. In the map, the KKH was subdivided into four different susceptibility zones. Some sections of the highway fall into high to very high susceptibility zones. According to results, active faults, slope gradient, seismicity and lithology have a strong influence on landslide events. Credibility of the map was validated by landslide density analysis (LDA) and receiver operator characteristics (ROC), yielding a predictive accuracy of 72 %, which is rated as satisfactory by previous researchers.

Application of Airborne LiDAR-Derived Parameters and Probabilistic-Based Frequency Ratio Model in Landslide Susceptibility Mapping

2012 ◽  
Vol 225 ◽  
pp. 442-447 ◽  
Author(s):  
Biswajeet Pradhan ◽  
Zulkiflee Abd. Latif ◽  
Siti Nur Afiqah Aman
Keyword(s):  
High Resolution ◽  
Landslide Susceptibility ◽  
Frequency Ratio ◽  
Susceptibility Mapping ◽  
Airborne Lidar ◽  
Landslide Susceptibility Mapping ◽  
Susceptibility Map ◽  
Ratio Model ◽  
Landslide Susceptibility Map ◽  
Frequency Ratio Model

The escalating number of occurrences of natural hazards such as landslides has raised a great interest among the geoscientists. Due to the extremely high number of point’s returns, airborne LiDAR permits the formation of more accurate DEM compared to other space borne and airborne remote sensing techniques. This study aims to assess the capability of LiDAR derived parameters in landslide susceptibility mapping. Due to frequent occurrence of landslides, Ulu Klang in Selangor state in Malaysia has been considered as application site. A high resolution of airborne LiDAR DEM was constructed to produce topographic attributes such as slope, curvature and aspect. These data were utilized to derive secondary deliverables of landslide parameters such as topographic wetness index (TWI), surface area ratio (SAR) and stream power index (SPI). A probabilistic based frequency ratio model was applied to establish the spatial relationship between the landslide locations and each landslide related factors. Subsequently, factor ratings were summed up to yield Landslide Susceptibility Index (LSI) and finally a landslide susceptibility map was prepared. To test the model performance, receiver operating characteristics (ROC) curve was carried out together with area under curve (AUC) analysis. The produced landslide susceptibility map demonstrated that high resolution airborne LiDAR data has huge potential in landslide susceptibility mapping.

scholarly journals Artificial intelligence approaches for spatial prediction of landslides in mountainous regions of western India

2021 ◽  
Author(s):  
PANKAJ PRASAD ◽  
Victor Joseph Loveson ◽  
Sumit Das ◽  
Priyankar Chandra
Keyword(s):  
Artificial Intelligence ◽  
Landslide Susceptibility ◽  
Landslide Susceptibility Mapping ◽  
Gradient Boosting ◽  
Susceptibility Map ◽  
Western India ◽  
Boosted Regression Tree ◽  
Study Region ◽  
Mountainous Regions ◽  
Stochastic Gradient Boosting

Abstract The prediction of landslide is a complex task but preparing the landslide susceptibility map through artificial intelligence approaches can reduce life loss and damages resulting from landslides. The purpose of this study is to evaluate and compare the landslide susceptibility mapping (LSM) using six machine learning models, including random forest (RF), deep boost (DB), stochastic gradient boosting (SGB), rotation forest (RoF), boosted regression tree (BRT), and logit boost (LB) in the mountainous regions of western India. The landslide inventory map consisting of 184 landslide locations has been divided into two groups for training (70% dataset) and validation (30% dataset) purposes. Fourteen landslide triggering factors including slope, topographical roughness index, road density, topographical wetness index, elevation, slope length, drainage density, stream power index, geomorphology, rainfall, soil, lithology, lineament density, and normalized difference vegetation index have been considered using the boruta approach for the LSM. The results reveal that the RF model has the highest precision in terms of AUC (0.88; 0.89), kappa (0.62; 0.50), accuracy (0.81; 0.77), and specificity (0.86; 0.86) both in the study region and secondary region, respectively. Hence, it can be concluded that the RF is an effective and promising technique as compared to DB, SGB, RoF, BRT, and LB for landslide susceptibility assessment in the research area as well as in regions having similar geo-environmental configuration.

scholarly journals GIS based Landslides Susceptibility Mapping using Probabilistic Certainty Factor for Hazard Zonation in Mandakini Valley

2020 ◽  
Vol 8 (6) ◽  
pp. 398-404
Keyword(s):  
High Risk ◽  
Landslide Susceptibility ◽  
Catchment Area ◽  
Susceptibility Mapping ◽  
Landslide Susceptibility Mapping ◽  
Susceptibility Map ◽  
Hazard Zonation ◽  
Certainty Factor ◽  
Landslide Susceptibility Map ◽  
Risk Zonation

Landslides are highly threatening a phenomenon which is very common in hilly region and mountainous regions. These landslides trigger major risks leading to heavy losses in terms of life and property. Many studies were conducted globally to determine Landslide vulnerability of different locations. In order to assess vulnerability, there were few studies around Landslides Susceptibility mapping also whose main objective is to identify high-risk vulnerable areas, there by applying measure to reduce the damage caused, if it were to happen in near future. In literature, there are many methods available for predictive susceptibility mapping of landslides. However, identification of any of the prevalent method for a specific area require utmost care and prudence because land sliding is a result of complex geo-environmental spatial factors. Mandakini valley is highly ruggedized terrain with intensive rains during monsoon season. As a result, Landslides are very common in the Mandakini River valley and its catchment area. These landslides cause severe damage to human settlements and infrastructure present in this area. In this study, we have used certainty factor method in order to generate landslide susceptibility map for the catchment area of Mandakini river. Certainty factor approach is a bi-variate probabilistic method which uses Geo-environmental parameters like elevation, slope, aspect, rainfall distance away from river, soil characteristics etc. to generate landslide susceptibility map. A Script was developed in ArcPy - a python package to design tools for generating susceptibility map. These tools can run both at desktop level and at server level and generate results in an integrated way. Esri ArcMap 10.7 is used in order to generate required data layers and thematic maps. Overall, this paper leverages GIS technology and its tools to performs Landslide Susceptibility Mapping using Probabilistic Certainty Factor and generate Hazard Zonation of Mandakini Valley using an automated script for generating Landslide Susceptibility Mapping and Hazard Risk Zonation. It was found that out of 696, total 136 villages are under high risk of landsides, total 329 villages are under moderate risks and around 231 villages are under low risk zonation impacting lives of approx. 216166 people. Also, it is worth mentioning that a GIS based script was developed to automate generation of Landslide Susceptibility Maps which can be used where the same geological and topographical feature prevails.

scholarly journals Application of Bayesian Hyperparameter Optimized Random Forest and XGBoost Model for Landslide Susceptibility Mapping

2021 ◽  
Vol 9 ◽  
Author(s):  
Shibao Wang ◽  
Jianqi Zhuang ◽  
Jia Zheng ◽  
Hongyu Fan ◽  
Jiaxu Kong ◽  
...  
Keyword(s):  
Random Forest ◽  
Decision Tree ◽  
Landslide Susceptibility ◽  
Susceptibility Mapping ◽  
Landslide Susceptibility Mapping ◽  
Gradient Boosting ◽  
The Loess Plateau ◽  
Tree Model ◽  
Validation Data ◽  
Extreme Gradient Boosting

Landslides are widely distributed worldwide and often result in tremendous casualties and economic losses, especially in the Loess Plateau of China. Taking Wuqi County in the hinterland of the Loess Plateau as the research area, using Bayesian hyperparameters to optimize random forest and extreme gradient boosting decision trees model for landslide susceptibility mapping, and the two optimized models are compared. In addition, 14 landslide influencing factors are selected, and 734 landslides are obtained according to field investigation and reports from literals. The landslides were randomly divided into training data (70%) and validation data (30%). The hyperparameters of the random forest and extreme gradient boosting decision tree models were optimized using a Bayesian algorithm, and then the optimal hyperparameters are selected for landslide susceptibility mapping. Both models were evaluated and compared using the receiver operating characteristic curve and confusion matrix. The results show that the AUC validation data of the Bayesian optimized random forest and extreme gradient boosting decision tree model are 0.88 and 0.86, respectively, which showed an improvement of 4 and 3%, indicating that the prediction performance of the two models has been improved. However, the random forest model has a higher predictive ability than the extreme gradient boosting decision tree model. Thus, hyperparameter optimization is of great significance in the improvement of the prediction accuracy of the model. Therefore, the optimized model can generate a high-quality landslide susceptibility map.

scholarly journals A Novel Intelligence Approach of a Sequential Minimal Optimization-Based Support Vector Machine for Landslide Susceptibility Mapping

2019 ◽  
Vol 11 (22) ◽  
pp. 6323 ◽  
Author(s):  
Pham ◽  
Prakash ◽  
Chen ◽  
Ly ◽  
Ho ◽  
...  
Keyword(s):  
Support Vector Machine ◽  
Hybrid Model ◽  
Landslide Susceptibility ◽  
Characteristic Curve ◽  
Susceptibility Mapping ◽  
Landslide Susceptibility Mapping ◽  
Support Vector ◽  
Sequential Minimal Optimization ◽  
Affecting Factors ◽  
Landslide Modeling

The main objective of this study is to propose a novel hybrid model of a sequential minimal optimization and support vector machine (SMOSVM) for accurate landslide susceptibility mapping. For this task, one of the landslide prone areas of Vietnam, the Mu Cang Chai District located in Yen Bai Province was selected. In total, 248 landslide locations and 15 landslide-affecting factors were selected for landslide modeling and analysis. Predictive capability of SMOSVM was evaluated and compared with other landslide models, namely a hybrid model of the cascade generalization optimization-based support vector machine (CGSVM), individual models, such as support vector machines (SVM) and naïve Bayes trees (NBT). For validation, different quantitative criteria such as statistical based methods and area under the receiver operating characteristic curve (AUC) technique were used. Results of the study show that the SMOSVM model (AUC = 0.824) has the highest performance for landslide susceptibility mapping, followed by CGSVM (AUC = 0.815), SVM (AUC = 0.804), and NBT (AUC = 0.800) models, respectively. Thus, the proposed novel SMOSVM model is a promising method for better landslide susceptibility mapping and prediction, which can be applied also in other landslide prone areas.

Sign in / Sign up

Export Citation Format

Share Document