Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning technique
AbstractIn recent years, several catastrophic landslide events have been observed throughout the globe, threatening to lives and infrastructures. To minimize the impact of landslides, the need of landslide susceptibility map is important. The study aims to extract high-quality non-landslide samples...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-01-01
|
| Series: | Geocarto International |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/10106049.2024.2327463 |
| _version_ | 1797238995456884736 |
|---|---|
| author | Chao Zhou Yue Wang Ying Cao Ramesh P. Singh Bayes Ahmed Mahdi Motagh Yang Wang Ling Chen Guangchao Tan Shanshan Li |
| author_facet | Chao Zhou Yue Wang Ying Cao Ramesh P. Singh Bayes Ahmed Mahdi Motagh Yang Wang Ling Chen Guangchao Tan Shanshan Li |
| author_sort | Chao Zhou |
| collection | DOAJ |
| description | AbstractIn recent years, several catastrophic landslide events have been observed throughout the globe, threatening to lives and infrastructures. To minimize the impact of landslides, the need of landslide susceptibility map is important. The study aims to extract high-quality non-landslide samples and improve the accuracy of landslide susceptibility modelling (LSM) outcomes by applying a coupled method of ensemble learning and Machine Learning (ML). The Zigui-Badong section of the Three Gorges Reservoir area (TGRA) in China was considered in the present study. Twelve influencing factors were selected as inputs for LSM, and the relationship between each causal factor and landslide spatial development was quantitatively analyzed. A total of 179 landslides have been used in the present study. About 70% of the landslide pixels were randomly considered for training, and the remaining 30% were used for validation. Logistic Regression (LR) model was applied to produce an initial susceptibility map, and the non-landslide samples were selected within the classified low-susceptibility zone. Subsequently, two ML classifiers – the Classification and Regression Tree (CART), and the Multi-Layer Perceptron (MLP), and four coupling models – the CART-Bagging, CART-Boosting, MLP-Bagging, and MLP-Boosting, were utilized for LSM. Finally, the receiver operating characteristics (ROC) curve and statistical analysis were applied for accuracy assessment. The results show that altitude and distance to rivers were the main causal factors of landslides in the study area. The LR-MLP-Boosting performed the best with an accuracy of 0.986 followed by the LR-CART-Bagging, LR-CART-Boosting, and LR-MLP-Bagging. Accuracy comparisons demonstrate that ensemble learning algorithm can notably enhance the LSM performance of ML classifiers, and the Boosting algorithm marginally outperforms the Bagging algorithm. Moreover, the LR model can effectively constrain the selection range of non-landslide samples. The non-landslide sampling method constrained by LR yields higher quality samples compared to raditional random sampling method with no constraints, which develops a more excellent LSM. |
| first_indexed | 2024-04-24T17:44:30Z |
| format | Article |
| id | doaj.art-ed98b677a0a44f25acefb9fe9d1d7499 |
| institution | Directory Open Access Journal |
| issn | 1010-6049 1752-0762 |
| language | English |
| last_indexed | 2024-04-24T17:44:30Z |
| publishDate | 2024-01-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Geocarto International |
| spelling | doaj.art-ed98b677a0a44f25acefb9fe9d1d74992024-03-27T14:19:40ZengTaylor & Francis GroupGeocarto International1010-60491752-07622024-01-0139110.1080/10106049.2024.2327463Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning techniqueChao Zhou0Yue Wang1Ying Cao2Ramesh P. Singh3Bayes Ahmed4Mahdi Motagh5Yang Wang6Ling Chen7Guangchao Tan8Shanshan Li9School of Geography and Information Engineering, China University of Geosciences, Wuhan, ChinaEngineering Faculty, China University of Geosciences, Wuhan, ChinaEngineering Faculty, China University of Geosciences, Wuhan, ChinaSchool of Life and Environmental Sciences, Schmid College of Science and Technology, Chapman University, Orange, CA, USAInstitute for Risk and Disaster Reduction, University College London (UCL), London, UKDepartment of Geodesy, Section of Remote Sensing and Geoinformatics, Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences, Potsdam, GermanyEngineering Faculty, China University of Geosciences, Wuhan, ChinaEngineering Faculty, China University of Geosciences, Wuhan, ChinaHydrogeology and Engineering Geology Institute of Hubei Geological Bureau, Jingzhou, ChinaHydrogeology and Engineering Geology Institute of Hubei Geological Bureau, Jingzhou, ChinaAbstractIn recent years, several catastrophic landslide events have been observed throughout the globe, threatening to lives and infrastructures. To minimize the impact of landslides, the need of landslide susceptibility map is important. The study aims to extract high-quality non-landslide samples and improve the accuracy of landslide susceptibility modelling (LSM) outcomes by applying a coupled method of ensemble learning and Machine Learning (ML). The Zigui-Badong section of the Three Gorges Reservoir area (TGRA) in China was considered in the present study. Twelve influencing factors were selected as inputs for LSM, and the relationship between each causal factor and landslide spatial development was quantitatively analyzed. A total of 179 landslides have been used in the present study. About 70% of the landslide pixels were randomly considered for training, and the remaining 30% were used for validation. Logistic Regression (LR) model was applied to produce an initial susceptibility map, and the non-landslide samples were selected within the classified low-susceptibility zone. Subsequently, two ML classifiers – the Classification and Regression Tree (CART), and the Multi-Layer Perceptron (MLP), and four coupling models – the CART-Bagging, CART-Boosting, MLP-Bagging, and MLP-Boosting, were utilized for LSM. Finally, the receiver operating characteristics (ROC) curve and statistical analysis were applied for accuracy assessment. The results show that altitude and distance to rivers were the main causal factors of landslides in the study area. The LR-MLP-Boosting performed the best with an accuracy of 0.986 followed by the LR-CART-Bagging, LR-CART-Boosting, and LR-MLP-Bagging. Accuracy comparisons demonstrate that ensemble learning algorithm can notably enhance the LSM performance of ML classifiers, and the Boosting algorithm marginally outperforms the Bagging algorithm. Moreover, the LR model can effectively constrain the selection range of non-landslide samples. The non-landslide sampling method constrained by LR yields higher quality samples compared to raditional random sampling method with no constraints, which develops a more excellent LSM.https://www.tandfonline.com/doi/10.1080/10106049.2024.2327463Reservoir landslidessusceptibility mappingnon-landslide samplingensemble learningmachine learning |
| spellingShingle | Chao Zhou Yue Wang Ying Cao Ramesh P. Singh Bayes Ahmed Mahdi Motagh Yang Wang Ling Chen Guangchao Tan Shanshan Li Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning technique Geocarto International Reservoir landslides susceptibility mapping non-landslide sampling ensemble learning machine learning |
| title | Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning technique |
| title_full | Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning technique |
| title_fullStr | Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning technique |
| title_full_unstemmed | Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning technique |
| title_short | Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning technique |
| title_sort | enhancing landslide susceptibility modelling through a novel non landslide sampling method and ensemble learning technique |
| topic | Reservoir landslides susceptibility mapping non-landslide sampling ensemble learning machine learning |
| url | https://www.tandfonline.com/doi/10.1080/10106049.2024.2327463 |
| work_keys_str_mv | AT chaozhou enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT yuewang enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT yingcao enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT rameshpsingh enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT bayesahmed enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT mahdimotagh enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT yangwang enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT lingchen enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT guangchaotan enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique AT shanshanli enhancinglandslidesusceptibilitymodellingthroughanovelnonlandslidesamplingmethodandensemblelearningtechnique |