Objective To investigate the suitability of different machine learning models and input variables for assessing collapsing gully susceptibility. Methods Taking Shicheng County, Jiangxi Province as the study area, an indicator system was constructed using geodetector (GD) for factor screening. The original values, frequency ratio (FR), and neighborhood frequency ratio (NFR) were used as input variables for the multilayer perceptron (MLP) and random forest (RF) models. The adaptability of these different models and input variables for Benggang susceptibility assessment was studied. Results 1) The AUC values of susceptibility assessment results from MLP and RF models under the NFR input variables were 0.854 and 0.860, respectively. Both models demonstrated good assessment performance, indicating that NFR was a suitable input variable. 2) The RF model generally outperformed the MLP model. Specifically, the Benggang densities in high susceptibility areas of original value-RF, NFR-RF, and FR-RF models were 3.93, 3.83, and 3.69, respectively. The original value-RF model demonstrated the strongest capability in identifying extremely high and high susceptibility areas. 3) The Benggang density was highest in the extremely high susceptibility area. Both high and extremely high susceptibility areas were concentrated in the northwest, closely matching the actual distribution pattern of Benggang. Conclusion NFR is a highly generalizable input variable. Compared with original values and FR, NFR exhibits the highest robustness in both MLP and RF models. The RF model is more suitable than the MLP model for assessing Benggang susceptibility.
WENH, NIS M, FENGS Y, et al. Effects of developmental stages and parts of collapsing gully on soil hydraulic properties in southern Jiangxi[J].Transactions of the Chinese Society of Agricultural Engineering,2019,35(24):136-143.
LIAOY S, TANGC Y, YUANZ J, et al. Research progress on collapsing gully erosion and its control in the red soil region of southern China[J].Acta Pedologica Sinica,2018,55(6):1297-1312.
DENGY S, CAIC F. Investigation and monitoring and control technology progress of collapsing gully erosion in red soil hilly region[J]. Acta Pedologica Sinica,2025,62(2):322-333.
PANF, WENH, YUANF, et al. Comparative assessment of Benggang development risk in Jiangxi Province based on information value model and frequency ratio model[J].Transactions of the Chinese Society of Agricultural Engineering,2025,41(5):107-115.
XIAL, XIAW X, GUOF, et al. Study on the influence of raster units with different spatial resolutions on the susceptibility of collapsing gully under subjective and objective weighting methods[J].Research of Soil and Water Conservation, 2025,32(6):208-215.
ZHANGX C, JIANGY L, WANGY J, et al. Landslide susceptibility assessment based on multi-objective optimization method[J].Journal of Soil and Water Conservation,2024,38(1):104-112.
[13]
ADNANM S G, RAHMANM S, AHMEDN, et al. Improving spatial agreement in machine learning-based landslide susceptibility mapping[J].Remote Sensing,2020,12(20):e3347.
[14]
ARABAMERIA, SAHAS, ROYJ, et al. Landslide susceptibility evaluation and management using different machine learning methods in the Gallicash River Watershed, Iran[J].Remote Sensing,2020,12(3):e475.
GUOF, LAIP, HUANGF M, et al. Literature review and research progress of landslide susceptibility evaluation based on knowledge graph[J].Earth Science,2024,49(5):1584-1606.
ZHAOY, CHENL X, FUS. Influence of variable value methods on collapse susceptibility[J]. Geospatial Information,2021,19(12):12-17.
[19]
BUIDT, TSANGARATOSP, NGUYENVT, et al. Comparing the prediction performance of a Deep Learning Neural Network model with conventional machine learning models in landslide susceptibility assessment[J].Catena,2020,188:e104426.
[20]
PHAMBT, PRAKASHI, BUIDT. Spatial prediction of landslides using a hybrid machine learning approach based on random subspace and classification and regression trees[J].Geomorphology,2018,303:256-270.
HUY, ZHANGZ Z, LINS H. Landslide susceptibility evaluation in Ili River valley, Xinjiang based on coupling of weights of evidence and logistic regression[J].Journal of Engineering Geology,2023,31(4):1350-1363.
CHENL H, LIL F, WUF, et al. Evaluation of geological hazard susceptibility in Beiliu City based on GIS and information value method[J].Earth and Environment, 2020,48(4):471-479.
GUOF, LAIP, CHENY, et al. Influence of different environmental factor connection methods on susceptibility evaluation of collapsing gully[J].Bulletin of Soil and Water Conservation, 2022,42(5):123-130.
SHENGM Q, LIUZ X, ZHANGX Q, et al. Landslide susceptibility prediction based on frequency ratio connection method and support vector machine[J].Science Technology and Engineering,2021,21(25):10620-10628.
KONGJ X, ZHUANGJ Q, PENGJ B, et al. Landslide susceptibility evaluation on the Loess Plateau based on information value and convolutional neural network[J].Earth Science, 2023,48(5):1711-1729.
YANGW L, BAIG S, SUNB, et al. Sensitivity of geological hazard factors and susceptibility evaluation in Ailaoshan area, Yunnan Province: A case study of Xinping County[J].The Chinese Journal of Geological Hazard and Control,2024,35(6):145-152.
BAIG S, YANGX M, ZHUJ Y, et al. Evaluation of geological hazard susceptibility in Wuhua District of Kunming based on weights of evidence method[J].The Chinese Journal of Geological Hazard and Control,2022,33(5):128-138.
LIC M, XUG L, LUY. Study on key influencing factors and sensitivity of collapsing gully in southeast Guangxi[J].Journal of Yangtze River Scientific Research Institute,2020,37(3):131-136.
[37]
WEIYJ. Identification of geo-environmental factors on Benggang susceptibility and its spatial modelling using comparative data-driven methods[J].Soil and Tillage Research,2021,208:e104857.
[38]
GAYENA, POURGHASEMIHR, SAHAS, et al. Gully erosion susceptibility assessment and management of hazard prone areas in India using different machine learning algorithms[J].Science of the Total Environment,2019,668:124-138.
SUNK, CHENGD B, HEJ J, et al. Comparative study on risk assessment methods of collapsing gully erosion: A case study of Guangdong Province[J].Soil and Water Conservation in China,2018(3):51-54.
[41]
LIUZ, WEIYJ, CUITT, et al. Spatial scaling effects of gully erosion in response to driving factors in southern China[J].Journal of Geographical Sciences,2024,34(5):942-962.
GUANJ L, HUANGY H, LINJ S, et al. Comparative assessment of collapsing gully risk based on information value model and random forest model[J].Mountain Research,2021,39(4):539-551.
YANGS, LID Y, YANL X, et al. Landslide susceptibility evaluation of high and steep bank slopes in Wujiang River based on random forest model[J]. Safety and Environmental Engineering, 2021, 28(4): 131-138.
LIUY Y, DIB F, ZHANY, et al. Debris flow susceptibility evaluation based on random forest model: A case study of severe disaster areas in Wenchuan earthquake[J].Mountain Research,2018,36(5):765-773.
[48]
CHENY. Spatial prediction and mapping of landslide susceptibility using machine learning models[J].Natural Hazards,2025,121(7):1-19.
CAOW G, PAND, XUZ J, et al. Research on landslide susceptibility mapping in Henan Province: Comparison of multiple machine learning models[J].Geological Science and Technology Bulletin,2025,44(1):101-111.
GUOF, ZHANGJ X, SHANH P, et al. Study on susceptibility evaluation of collapsing gully in Shicheng County, Ganzhou City under different grid resolutions[J].Journal of China Three Gorges University (Natural Sciences),2024,46(2):63-70.
[53]
HONGH, LIUJ, BUIDT, et al. Landslide susceptibility mapping using J48 Decision Tree with AdaBoost, Bagging and Rotation Forest ensembles in the Guangchang area (China)[J].Catena,2018,163:399-413.
[54]
LIL, LANH, GUOC, et al. A modified frequency ratio method for landslide susceptibility assessment[J].Landslides,2017,14:727-741.
[55]
ZHANGY, LANH, LIL, et al. Optimizing the frequency ratio method for landslide susceptibility assessment: A case study of the Caiyuan basin in the southeast mountainous area of China[J]. Journal of Mountain Science, 2020, 17(2): 340-357.
ZHOUP, DENGH, ZHANGW J, et al. Landslide susceptibility evaluation based on information value model and machine learning methods: A case study of Lixian County, Sichuan[J].Scientia Geographica Sinica,2022,42(9):1665-1675.
[58]
SEVGENE, KOCAMANS, NEFESLIOGLUH A, et al. A novel performance assessment approach using photogrammetric techniques for landslide susceptibility mapping with logistic regression, ANN and Random Forest[J].Sensors,2019,19(18):e3940.
WUX Q, LAIC G, CHENX H, et al. Landslide hazard evaluation based on random forest weights: A case study of Dongjiang River basin[J].Journal of Natural Disasters,2017,26(5):119-129.
WANGS B, ZHUANGJ Q, FANH Y, et al. Landslide susceptibility evaluation based on frequency ratio and ensemble learning: A case study of Batang-Dege section in the upper reaches of Jinsha River[J].Journal of Engineering Geology,2022,30(3):817-828.
ZHAIW H, WANGX D, WUM T, et al. Evaluation of geological hazard susceptibility based on coupling of frequency ratio model and random forest model[J]. Journal of Natural Disasters, 2023, 32(6): 74-82.
WANGX W, ZHANGL L, MOD K, et al. Evaluation of susceptibility of slope geological hazards in Pingguo City based on coupling of information value and multilayer perceptron classifier model[J].Carsologica Sinica,2023,42(2):370-381.
[67]
王劲峰,徐成东.地理探测器:原理与展望[J].地理学报,2017,72(1):116-134.
[68]
WANGJ F, XUC D. Geodetector: Principle and prospect[J].Acta Geographica Sinica,2017,72(1):116-134.
GUOF, JIANGG H, HUANGX H, et al. The influence of environmental factor combination and negative sample selection strategy on the susceptibility evaluation of rockfall in granite areas [J].Transactions of the Chinese Society of Agricultural Engineering,2024,40(1):191-200.