Landslide susceptibility assessment using locally weighted learning integrated with machine learning algorithms

Expert Systems with Applications, Год журнала: 2023, Номер 237, С. 121678 - 121678

Опубликована: Сен. 19, 2023

Язык: Английский

---

Enhancing landslide susceptibility modelling through a novel non-landslide sampling method and ensemble learning technique

Geocarto International, Год журнала: 2024, Номер 39(1)

Опубликована: Янв. 1, 2024

In recent years, several catastrophic landslide events have been observed throughout the globe, threatening to lives and infrastructures. To minimize impact of landslides, need susceptibility map is important. The study aims extract high-quality non-landslide samples improve accuracy modelling (LSM) outcomes by applying a coupled method ensemble learning Machine Learning (ML). Zigui-Badong section Three Gorges Reservoir area (TGRA) in China was considered present study. Twelve influencing factors were selected as inputs for LSM, relationship between each causal factor spatial development quantitatively analyzed. A total 179 landslides used About 70% pixels randomly training, remaining 30% validation. Logistic Regression (LR) model applied produce an initial map, within classified low-susceptibility zone. Subsequently, two ML classifiers – Classification Tree (CART), Multi-Layer Perceptron (MLP), four coupling models CART-Bagging, CART-Boosting, MLP-Bagging, MLP-Boosting, utilized LSM. Finally, receiver operating characteristics (ROC) curve statistical analysis assessment. results show that altitude distance rivers main area. LR-MLP-Boosting performed best with 0.986 followed LR-CART-Bagging, LR-CART-Boosting, LR-MLP-Bagging. Accuracy comparisons demonstrate algorithm can notably enhance LSM performance classifiers, Boosting marginally outperforms Bagging algorithm. Moreover, LR effectively constrain selection range samples. sampling constrained yields higher quality compared raditional random no constraints, which develops more excellent

Язык: Английский

---

Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment

Geo-spatial Information Science, Год журнала: 2025, Номер unknown, С. 1 - 21

Опубликована: Янв. 17, 2025

To address the errors of negative samples in landslide susceptibility modeling and traditional methods exploring regularities hidden evaluation factors, this paper proposes a stacking one- three-dimensional Convolutional Neural Network (Stacking-1D-3D-CNN) assessment method considering sample optimization selection. First, order to select rationally, adopts Relative Frequency Ratio combined with Certainty Factor Method (RFR-CFM) determine samples; secondly, Stacking-1D-3D-CNN proposed is RFR-CFM for first time assessment. In work, determined by Information Quality Model (IQM) were historical disaster points form total sample, modeled at different ratios. Finally, it compared several other models terms hazard zoning results, prone zone statistics, model performance. The findings show that degree spatial aggregation training testing has much greater impact on accuracy than their proportions. Furthermore, models, RFR-CFM-Stacking-1D-3D-CNN highest AUC value, precision, recall, F-score, accuracy, which are 0.95, 0.83, 0.89, 0.85, 84.76%, respectively, lowest RMSE MAE, 0.39 0.15, respectively. This proves selection method's rationality model's effectiveness.

Язык: Английский

---

Comprehensive review of remote sensing integration with deep learning in landslide forecasting and future directions

Natural Hazards, Год журнала: 2025, Номер unknown

Опубликована: Март 10, 2025

Язык: Английский

---

Landslide susceptibility mapping using multiple combination weighting determination: a case study of collector roads in Pingshan County, Hebei Province, China

Frontiers in Earth Science, Год журнала: 2024, Номер 12

Опубликована: Июль 10, 2024

The landslide susceptibility map estimates the quantitative relationship between known landslides and control factors, it has been used for site selection of infrastructures geo-disaster management. As rockfalls occur frequently in mountainous areas Hebei Province, China, due to road construction, managing government needs evaluate vulnerability geo-disasters slopes avoid unfavorable subsequent constructions. Some typical collector were as study area Pingshan County, Province. By analyzing triggering we determined classification criteria proposed a comprehensive method determining weighting. respective weighting was calculated by AHP CRITIC method, combination game theory method. roads evaluated mapped using ArcGIS platform. validated field investigation. validation results show effectiveness methods, given good number correctly classified landslides. could have significant impact on reducing infrastructure China.

Язык: Английский

---

An Investigation into the Susceptibility to Landslides Using Integrated Learning and Bayesian Optimization: A Case Study of Xichang City

Sustainability, Год журнала: 2024, Номер 16(20), С. 9085 - 9085

Опубликована: Окт. 20, 2024

In the middle southern section of Freshwater River–Small River Fault system, Xichang City, Daliang Prefecture, Sichuan Province, is situated in junction between Anning and Zemu Fault. There has been a risk increased activity fault zone recent years, landslide susceptibility evaluation for area can effectively reduce disaster occurrence. Using integrated learning Bayesian hyperparameter optimization, 265 landslides City were used as samples this study. Thirteen influencing factors chosen to assess susceptibility, BO-XGBoost, BO-LightGBM, BO-RF models evaluated using precision, recall, F1, accuracy, AUC curves. The findings indicated that after removing terrain relief factor, four most significant associated with NDVI, distance from faults, slope, rivers. study demonstrates value BO-XGBoost model 0.8677, demonstrating better generalization ability higher prediction accuracy than BO-LightGBM models. After optimization hyperparameters, offers improvement accuracy.

Язык: Английский

---

Predicting the Spatial Distribution of Geological Hazards in Southern Sichuan, China, Using Machine Learning and ArcGIS

Land, Год журнала: 2025, Номер 14(3), С. 577 - 577

Опубликована: Март 10, 2025

Geological hazards in Southern Sichuan have become increasingly frequent, posing severe risks to local communities and infrastructure. This study aims predict the spatial distribution of potential geological using machine learning models ArcGIS-based analysis. A dataset comprising 2700 known hazard locations Yibin City was analyzed extract key environmental topographic features influencing susceptibility. Several were evaluated, including random forest, XGBoost, CatBoost, with model optimization performed Sparrow Search Algorithm (SSA) enhance prediction accuracy. produced high-resolution susceptibility maps identifying high-risk zones, revealing a distinct pattern characterized by concentration mountainous areas such as Pingshan County, Junlian Gong while plains exhibited relatively lower risk. Among different types, landslides found be most prevalent. The results further indicate strong overlap between predicted zones existing rural settlements, highlighting challenges resilience these areas. research provides refined methodological framework for integrating geospatial analysis prediction. findings offer valuable insights land use planning mitigation strategies, emphasizing necessity adopting “small aggregations multi-point placement” approach settlement Sichuan’s regions.

Язык: Английский

---

Mapping landslide susceptibility in the Eastern Mediterranean mountainous region: a machine learning perspective

Environmental Earth Sciences, Год журнала: 2025, Номер 84(9)

Опубликована: Апрель 30, 2025

Язык: Английский

---

Landslide spatial prediction using data-driven based statistical and hybrid computational intelligence algorithms

Geocarto International, Год журнала: 2025, Номер 40(1)

Опубликована: Май 26, 2025

Язык: Английский

---

Landslide susceptibility evaluation and determination of critical influencing factors in eastern Sichuan mountainous area, China

Ecological Indicators, Год журнала: 2024, Номер 169, С. 112911 - 112911

Опубликована: Ноя. 28, 2024

Язык: Английский

---