Assessment and Comparative Evaluation of Machine Learning Algorithms (SVM, ANN, and MaxEnt) for Landslide Susceptibility Zonation (LSZ) in the Ziarat Watershed, Golestan Province, Iran

Articles in Press

Document Type : Research Article

Authors

Habib Nazarnejad ¹

Aiding Kornejady ²

Chooqhi Bairam Komaki ¹

Farkhondeh Aghajan Liafu ¹

¹ Department of Watershed Management, Faculty of Range and Watershed Management, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran

² Golestan Agricultural and Natural Resources Research and Education Center, Gorgan, Iran

10.22067/geoeh.2026.97226.1635

Abstract

Landslide is one of the most destructive slope hazards in mountainous regions. The performance of three machine learning models — Artificial Neural Network (ANN), Support Vector Machine (SVM), and Maximum Entropy (MaxEnt) — was evaluated and compared for landslide susceptibility mapping in the Ziarat watershed (Golestan Province). A total of 116 landslide points were recorded through GPS field surveys. Stratified random sampling was applied to preserve the spatial distribution of points across geological units, elevation classes, and slope categories. Of these, 81 points (70%) were allocated for training and 35 points (30%) for validation. A total of 5 main factor groups (topography, hydrology, geology, morphometry, and anthropogenic factors) comprising 23 sub-factors were selected: slope, aspect, elevation, valley depth, profile and plan curvature, topographic position index, relative slope position, convergence index, positive and negative openness, rainfall, distance from drainage network, topographic wetness index, stream power index, flow accumulation, geological formation, distance from faults, soil texture, mass balance index, Melton ruggedness index, and vector ruggedness measure. Modeling was performed using ModEco software, and accuracy assessment was carried out using the ROC curve and the Area Under the Curve (AUC) index. Results showed that the ANN model (AUC = 0.936) and the SVM model (AUC = 0.934) performed comparably well, while the MaxEnt model (AUC = 0.904) ranked lower. The most influential factors controlling landslide occurrence were, in order: elevation (corresponding to zones with annual rainfall exceeding 550 mm), the Shemshak and Gorgan Schist formations, and distance from residential areas. Given the ongoing land use changes and increasing construction activity in the Ziarat watershed, restrictions on land use conversion and a ban on construction on steep slopes are recommended.
Introduction
Landslides represent a prevalent and destructive geohazard globally, imposing substantial socioeconomic costs through loss of life, infrastructure damage, and environmental degradation. Their recurrent nature and profound destructive potential underscore the imperative for advanced research in watershed management and geotechnical engineering to devise robust predictive and mitigation strategies. Landslide susceptibility zonation (LSZ) stands as a critical spatial planning and management tool, enabling the precise identification of areas inherently prone to slope instability. This proactive approach facilitates informed land-use planning and targeted risk reduction interventions. The intrinsic complexity and non-linear interactions among diverse geomorphological, hydrological, and anthropogenic factors governing landslide initiation frequently render conventional modeling techniques insufficient. Consequently, the application of sophisticated machine learning algorithms has gained considerable traction in LSZ studies, attributed to their enhanced capacity to model these intricate spatial relationships. Selecting an accurate and computationally efficient model is crucial for reliable landslide prediction and effective risk management in watershed planning. This research endeavors to rigorously assess and comparatively analyze the predictive performance of three prominent machine learning models—Artificial Neural Network (ANN), Support Vector Machine (SVM), and Maximum Entropy (MaxEnt)—in generating accurate landslide susceptibility maps for the Ziarat watershed, located in Golestan Province, Iran. The primary objective is to ascertain the optimal predictive model for this geomorphologically dynamic region, thereby contributing to more effective landslide hazard management.
Material and Methods
The study area, the Ziarat watershed in Golestan Province, Iran, is characterized by diverse topography and geological formations, rendering it susceptible to frequent landslide events. The initial step in landslide susceptibility zoning involves generating a spatial distribution map of historical landslide events, which serves as a critical foundation for subsequent susceptibility and hazard assessment studies. A comprehensive landslide inventory was compiled through detailed field investigations, identifying 116 distinct landslide locations. To ensure model impartiality and robust validation, these locations were stochastically partitioned: 81 points (%70) were designated for model training, while the remaining 35 points (%30) were reserved for independent model validation. A total of 23 conditioning factors, recognized for their influence on landslide occurrence, were meticulously prepared as input variables. These factors encompassed a broad spectrum of environmental parameters, including eleven topographic indices (aspect, slope degree, elevation, valley depth, profile curvature, plan curvature, LS factor, relative slope position (RSP), convergence, positive openness, negative openness), five hydrological variables (rainfall, distance to drainage network, topographic wetness index (TWI), stream power index (SPI), flow accumulation), three geological attributes (Lithology, distance to faults, soil texture), and three morphometric indices (mass balance index (MBI), Melton roughness index (MRI), vector ruggedness measure (VRM)), along with one anthropogenic factor (distance to residential areas). To evaluate multicollinearity among predictive factors, tolerance and Variance Inflation Factor (VIF) indices were computed using SPSS version 26. The machine learning models (ANN, SVM, and MaxEnt) were implemented within the ModEco software environment, configured to elucidate the complex, non-linear relationships between these environmental factors and landslide susceptibility. Model performance was quantitatively evaluated using the Receiver Operating Characteristic (ROC) curve and the Area Under the Curve (AUC) metric.
Results and Discussion
The comparative assessment of the three machine learning models for landslide susceptibility mapping in the Ziarat watershed yielded distinct performance outcomes during the validation phase. The Artificial Neural Network (ANN) model, achieving an Area Under the Curve (AUC) value of 0.936 and the Support Vector Machine (SVM) model with an AUC value of 0.934 exhibited a highly comparable and robust performance. These results signify the exceptional accuracy and reliability of both ANN and SVM models in delineating landslide-prone areas within the study region. In contrast, the Maximum Entropy (MaxEnt) model, while demonstrating good performance, recorded a comparatively lower AUC of 0.904. The marginal difference in AUC between ANN and SVM suggests their analogous effectiveness in capturing the intricate spatial dependencies between the 23 conditioning factors and landslide incidence. Further analysis focused on quantifying the relative importance of these conditioning factors. This assessment revealed that elevation, geological characteristics, and distance from residential areas were the most influential predictors of landslide susceptibility. This finding is highly consistent with the local geomorphological context, where steep slopes at higher elevations and the presence of unstable geological formations are primary drivers of instability. The significant influence of distance from residential areas highlights the critical role of anthropogenic pressures in exacerbating landslide risk in the watershed.
Conclusion
This research successfully evaluated the efficacy of Artificial Neural Network (ANN), Support Vector Machine (SVM), and Maximum Entropy (MaxEnt) models for landslide susceptibility mapping in the Ziarat watershed. The generated susceptibility maps represent invaluable assets for local authorities and land-use planners, offering a scientific basis for targeted risk management and mitigation strategies. Specifically, these maps can inform the prioritization of slope stabilization measures and guide sustainable development in high-risk zones. This study significantly advances the understanding of landslide susceptibility modeling in complex terrains and offers a transferable methodology for similar investigations in other susceptible regions. For future research, it is recommended to employ hybrid modeling approaches that integrate machine learning algorithms with deep learning architectures to enhance the accuracy and robustness of landslide prediction. Furthermore, given the observed land-use transformations including the expansion of residential zones and increased villa construction in the Ziarat watershed implementing regulatory measures to constrain unplanned land-use changes in this area is strongly advised to mitigate landslide risk.

Keywords

Maximum Entropy

Landslide Susceptibility Zoning

Artificial Neural Network

Support Vector Machine

Machine Learning

Subjects

• Sliding and mass movements

---

©2026 The author(s). This is an open access article distributed under Creative Commons Attribution 4.0 International License (CC BY 4.0)

 

 

References

Ado, M., Amitab, K., Maji, A. K., Jasińska, E., Gono, R., Leonowicz, Z., & Jasiński, M. (2022). Landslide Susceptibility Mapping Using Machine Learning: A Literature Survey. Remote Sensing, 14(13), 3029.  https://doi.org/10.3390/rs14133029

Arabameri, A., Rezaei, K., Yamani, M., & Shirani, K. (2020). Performance assessment of integrated model of data driven EBF model and knowledge driven AHP model in Landslide spatial modeling (Case study: Ferydoun Shahr watershed). Journal of Water and Soil Conservation Research, 25(5), 47–67. [In Persian] https://doi.org/10.22069/jwsc.2019.14338.2907

Babarabi, N., Lorestani, Gh., & Esmaili. R. (2025). Landslide hazard zoning using random forest and support vector machine models (Case study: Talar basin). Researches in Earth Sciences, 16(1), 152-168. [In Persian] https://doi.org/10.48308/esrj.2025.236839.1234

Bahmani, S., Zangneh Tabar, Z., Mohammadi, S., & Matai, S. (2021). Identification and zoning of landslide prone areas using object-oriented method and network analysis (ANP) (Case study: Shahroudchai catchment area of Khalkhal city). Geography and Environmental Hazards ,10(3), 39-59. [In Persian] https://dx.doi.org/10.22067/geoeh.2021.67924.1006

Beven, K. J., & Kirkby, M. J. (1979). A physically based, variable contributing area model of basin hydrology. Hydrological Sciences Bulletin, 24(1), 43-69. https://doi.org/10.1080/02626667909491834

Boos, D. D., & Brownie, C. (2004). Comparing variances and other measures of dispersion. Statistical Science, 19, 571–578. https://doi.org/10.1214/088342304000000503

Chen, W., Pourghasemi, H. R., Kornejady, A., & Zhang, N. (2017). Landslide spatial modeling: introducing new ensembles of ANN, MaxEnt, and SVM machine learning techniques. Geoderma, 305, 314-327. https://doi.org/10.1016/j.geoderma.2017.06.020

Dastranj, A., & Karimi Sangchini, E. (2022). Landslide susceptibility predicting using the maximum entropy machine learning algorithm(Bar catchment of Nishapur). Earth Science Research Journal, 13(51), 79–96. [In Persian] https://doi.org/10.48308/esrj.2022.102965

Dastranj, A., Karimi Sangchini, E., & Noor, H. (2024).Landslide risk assessment in the Bar watershed, Neyshabur, Razavi Khorasan Province. Iranian Journal of Watershed Management Science and Engineering, 18(67), 44–60. [In Persian] https://doi.org/10.22034/18.67.4

Emadodin, S., Taheri, V., Mohamadghasemi, M., & Nazari, Z. (2021). Landslide Susceptibility Zonation applying frequency ratio models and statistical index in in Oghan watershed. Quantitative Geomorphological Research, 9(4), 75–95. [In Persian] https://doi.org/10.22034/gmpj.2021.248268.1211

Faramarzi, H., Hosseini, S. M., Pourghasemi, H. R., & Farnaghi, M. (2023). Using machine learning techniques in multi-hazards assessment of Golestan National Park, Iran. Natural Hazards, 117(3), 3231-3255. https://doi.org/10.1007/s11069-023-05984-1

Fathiganji, R., Nazari Samani, A. A., Feiznia, S., & Nourizdan, A. (2024). Evaluating the role of geomorphic characteristics in landslide vulnerability and sensitivity (Case study: Kakasharaf Drainage Basin, Lorestan Province. Natural Environment, 77(4), 731–747. [In Persian] http//doi.org/10.22059/jne.2024.376982.2675

Forte, G., Verrucci, L., Di Giulio, A., De Falco, M., Tommasi, P., Lanzo, G., ... & Santo, A. (2021). Analysis of major rock slides that occurred during the 2016–2017 Central Italy seismic sequence. Engineering Geology, 290, 106194.‏ https://doi.org/10.1016/j.enggeo.2021.106194

Froude, M. J., & Petley, D. N. (2018). Global fatal landslide occurrence from 2004 to 2016. Natural Hazards and Earth System Sciences, 18(8), 2161-2181. https://doi.org/10.5194/nhess-18-2161-2018

Gariano, S. L., & Guzzetti, F. (2016). Landslides in a changing climate. Earth-Science Reviews, 162, 227-252.‏ https://doi.org/10.1016/j.earscirev.2016.08.011

Ghanavati, E., Ahmadabadi, A., & Gholami, M. (2019). Landslide susceptibility mapping of Kan using index of Entropy and LSM. Quantitative Geomorphological Research, 8(1), 16–33. [In Persian] https:// doi.org/10.22067/geoeh.2025.89542.1514

Ghiasi, S. S., Rajabzadeh, F., Najirad, S.,  Feiznia, S., & Nazari Samani, A. A. (2017). Determining effective factors in the occurrence of shallow landslides in the  Khiov-chai watershed. Watershed Engineering and Management, 9(2), 140–154. [In Persian] https://doi.org/10.22092/ijwmse.2017.109697

Gilanipoor, A., Motevalli, S., & Derafshi, K. (2025). Assessment of landslide sensitivity and determination of effective factors in its occurrence using the random forest algorithm (Case study: Glandrood watershed). Journal of Geography and Environmental Hazards, 14(1), 247-274. [In Persian]  https://doi.org/10.22067/geoeh.2025.89542.1514

Guns, M., & Vanacker, V. (2014). Shifts in landslide frequency-area distribution after forest conversion in the tropical Andes. Anthropocene, 6, 75–85. https://doi.org/10.1016/j.ancene.2014.08.001

Hamidi, M., Bouramtane, T., Abraham, S., Kacimi, I., Barbiero, L., Kassou, N., ... & Levy, G. (2023). Landslide hazard assessment in the heterogeneous geomorphological and environmental context of the rif region, Morocco– A machine learning approach. Ecological Engineering & Environmental Technology, 24, 272-292. ‏ https://doi.org/10.12912/27197050/172569

Hanifinia, A. A., & Nazarnejad, H. (2022). The Effect of Morphometry Indices on Improving the Performance of Data Mining Models for Landslide Sensitivity Zoning in Cherikabad Watershed, Urmia.  Geography and Environmental Hazards,10(40), 47-68. [In Persian] https://dx.doi.org/10.22067/geoeh.2021.69707.1041

Harmouzi, H., Nefeslioglu, H. A., Rouai, M., Sezer, E. A., Dekayir, A., & Gokceoglu, C. (2019). Landslide susceptibility mapping of the Mediterranean coastal zone of Morocco between Oued Laou and El Jebha using artificial neural networks (ANN). Arabian Journal of Geosciences, 12(22), 696.‏ https://doi.org/10.1007/s12517-019-4892-0

Hemasinghe, H., Rangali, R. S. S., Deshapriya, N. L., & Samarakoon, L. (2018). Landslide susceptibility mapping using logistic regression model (a case study in Badulla District, Sri Lanka). Procedia Engineering, 212, 1046-1053. https://doi.org/10.1016/j.proeng.2018.01.135

Hong, H., Pradhan, B., Xu, C., & Bui, D. T. (2015). Spatial prediction of landslide hazard at the Yihuang area (China) using two-class kernel logistic regression, alternating decision tree and support vector machines. Catena, 133, 266-281.‏ https://doi.org/10.1016/j.catena.2015.05.019

Kalantar, B., Ueda, N., Saeidi, V., Ahmadi, K., Halin, A. A., & Shabani, F. (2020). Landslide susceptibility mapping: machine and ensemble learning based on remote sensing big data. Remote Sensing, 12(11), 1737. https://doi.org/10.3390/rs12111737

Karakas, G., Kocaman, S., & Gokceoglu, C. (2022). On the effect of DEM quality for landslide susceptibility mapping. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 3, 525-531.‏ https://doi.org/10.5194/isprs-annals-V-3-2022-525-2022

Karami, F., Bayati Khatibi, M., Kheirizadeh, M., & Mokhtari Asl, A. (2019). Evaluation of Performance of Support Vector Machine Algorithm in Landslide Susceptibility Zoning in Ahar-chai Basin. Geography and Environmental Hazards, 8(4), 1–17. [In Persian] https://doi.org/10.22067/geo.v8i4.83263  

Kirschbaum, D., Stanley, T., & Zhou, Y. (2015). Spatial and temporal analysis of a global landslide catalog. Geomorphology, 249, 4-15.‏ https://doi.org/10.1016/j.geomorph.2015.03.016

Kornejady, A. (2013). Landslide hazard, damage assessment, and management planning in the Ziarat watershed. (Master's Thesis). Gorgan University of Agricultural Sciences and Natural Resources, Iran. [In Persian]

Kornejady, A., & Pourghasemi, H. R. (2019). Landslide susceptibility assessment using data mining models, a case study: Chehel-Chai Basin. Watershed Engineering and Management, 11(1), 28-42. [In Persian] https://doi.org/10.22092/ijwmse.2019.118436

Kornejady, A., Ownegh, M., Pourghasemi, H. R., Bahremand, A., & Motamedi, M. (2020). Landslide susceptibility prediction using the coupled Mahalanobis distance and machine learning models (case study: Owghan watershed, Golestan province. Researches in Earth Science, 11(2), 1–18. [In Persian] https://doi.org/10.52547/esrj.11.2.1

Lee, S. A. R. O. (2005). Application of logistic regression model and its validation for landslide susceptibility mapping using GIS and remote sensing data. International Journal of Remote Sensing, 26(7), 1477-1491.‏ https://doi.org/10.1080/01431160412331331012

Maki Mateso, J. C., Bielders, C. L., Monsieurs, E., Depicker, A., Smets, B., Tambala, T., ... & Dewitte, O. (2023). Characteristics and causes of natural and human-induced landslides in a tropical mountainous region: the rift flank west of Lake Kivu (Democratic Republic of the Congo). Natural Hazards and Earth System Sciences, 23(2), 643-666. ‏ https://doi.org/10.5194/nhess-23-643-2023

Mao, Y., Li, Y., Teng, F., Sabonchi, A. K., Azarafza, M., & Zhang, M. (2024). Utilizing hybrid machine learning and soft computing techniques for landslide susceptibility mapping in a drainage basin. Water, 16(3), 380.‏ https://doi.org/10.3390/w16030380

Moavi, M., Elmizadeh, H., & Entezari, M. (2025). Landslide Susceptibility Modelling Using Artificial Neural Network Algorithm: A Case Study of the Shahid Abbaspour Dam Catchment, Northeastern Khuzestan. Journal of Geography and Environmental Hazards, 14(3), 38-54. https://doi.org/10.22067/geoeh.2025.92634.1557

Nanehkaran, Y. A., Chen, B., Cemiloglu, A., Chen, J., Anwar, S., Azarafza, M., & Derakhshani, R. (2023). Riverside landslide susceptibility overview: leveraging artificial neural networks and machine learning in accordance with the United Nations (UN) sustainable development goals. Water, 15(15), 2707.‏ https://doi.org/10.3390/w15152707

Nanehkaran, Y. A., Mao, Y., Azarafza, M., Kockar, M. K., & Zhu, H. H. (2021). Fuzzy-based multiple decision method for landslide susceptibility and hazard assessment: A case study of Tabriz, Iran. Geomechanics and Engineering, 24(5), 407-418. https://doi.org/10.12989/gae.2021.24.5.407

Nikoobakht, S., Azarafza, M., Akgün, H., & Derakhshani, R. (2022). Landslide susceptibility assessment by using convolutional neural network. Applied Sciences, 12(12), 5992.‏ https://doi.org/10.3390/app12125992

Ozturk, U., Bozzolan, E., Holcombe, E. A., Shukla, R., Pianosi, F., & Wagener, T. (2022). How climate change and unplanned urban sprawl bring more landslides. Nature, 608(7922), 262-265.‏ https://doi.org/10.1038/d41586-022-02141-9

Park, S., & Kim, J. (2019). Landslide susceptibility mapping based on random forest and boosted regression tree models, and a comparison of their performance. Applied Sciences, 9(5), 942. ‏ https://doi.org/10.3390/app9050942

Pawluszek, K., & Borkowski, A. (2017). Impact of DEM-derived factors and analytical hierarchy process on landslide susceptibility mapping in the region of Rożnów Lake, Poland. Natural Hazards, 86(2), 919-952.‏ https://doi.org/10.1007/s11069-016-2725-y

Petley, D. (2012). Global patterns of loss of life from landslides. Geology, 40(10), 927-930.‏ https://doi.org/10.1130/G33217.1

Pham, B. T., Pradhan, B., Bui, D. T., Prakash, I., & Dholakia, M. B. (2016). A comparative study of different machine learning methods for landslide susceptibility assessment: A case study of Uttarakhand area (India). Environmental Modelling & Software, 84, 240-250.‏ https://doi.org/10.1016/j.envsoft.2016.07.005

Phong, T. V., Phan, T. T., Prakash, I., Singh, S. K., Shirzadi, A., Chapi, K., ... & Pham, B. T. (2021). Landslide susceptibility modeling using different artificial intelligence methods: A case study at Muong Lay district, Vietnam. Geocarto International, 36(15), 1685-1708.‏ https://doi.org/10.1080/10106049.2019.1665715

Pontius Jr, R. G., & Schneider, L. C. (2001). Land-cover change model validation by an ROC method for the Ipswich watershed, Massachusetts, USA. Agriculture, Ecosystems & Environment, 85(1-3), 239-248. ‏ https://doi.org/10.1016/S01678809(01)00187-6

Pourghasemi, H. R., Pradhan, B., Gokceoglu, C., & Deylami Moezzi, K. (2012). Landslide susceptibility mapping using a spatial multi criteria evaluation model at Haraz Watershed, Iran. In Terrigenous mass movements: detection, modelling, early warning and mitigation using geoinformation technology (pp. 23-49). Berlin, Heidelberg: Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-25495-6_2

Pyakurel, A., Dahal, B. K., & Gautam, D. (2023). Does machine learning adequately predict earthquake induced landslides?. Soil Dynamics and Earthquake Engineering, 171, 107994. ‏ https://doi.org/10.1016/j.soildyn.2023.107994

Saberchenari, K., Salmani, H., & Mirabedini, M. (2018). Landslide hazard mapping using Dempster-Shafer theory - A case study: Ziarat Watershed, Golestan Province, Iran. Journal of Engineering Geology, 11(4), 385-404. ‎ https://doi.org/10.18869/acadpub.jeg.11.4.385

Sadati, S. H., Mousavi, S. R., Vahabzadeh Kebria, G., & Roshun, S. H. (2025). Evaluation of Random Forest and Support Vector Machine Models in Landslide Risk Mapping (Case study: Tajan Basin, Mazandaran Province). Journal of Natural Environmental Hazards, 14(45), 133-154. https://doi.org/10.22111/jneh.2025.50031.2071

Santangelo, N., Forte, G., De Falco, M., Chirico, G. B., & Santo, A. (2021). New insights on rainfall triggering flow-like landslides and flash floods in Campania (Southern Italy). Landslides, 18(8), 2923-2933.‏ https://doi.org/10.1007/s10346-021-01667-9

Saro, L., Woo, J. S., Kwan-Young, O., & Moung-Jin, L. (2016). The spatial prediction of landslide susceptibility applying artificial neural network and logistic regression models: A case study of Inje, Korea. Open Geosciences, 8(1), 117-132. ‏ https://doi.org/10.1515/geo-2016-0010

Selamat, S. N., Abd Majid, N., & Mohd Taib, A. (2023). A comparative assessment of sampling ratios using artificial neural network (ANN) for landslide predictive model in Langat River Basin, Selangor, Malaysia. Sustainability, 15(1), 861.‏ https://doi.org/10.3390/su15010861

Sepahvand, A. R., & Beiranvand, N. (2024). Landslide susceptibility mapping using various soft computing techniques (Case study: A part of Haraz Watershed). Water and Soil Management and Modelling, 4(2), 261-278. https://doi.org/10.22098/mmws.2023.12678.1263

Sun, D., Xu, J., Wen, H., & Wang, D. (2021). Assessment of landslide susceptibility mapping based on Bayesian hyperparameter optimization: A comparison between logistic regression and random forest. Engineering Geology, 281, 105972.‏ https://doi.org/10.1016/j.enggeo.2020.105972

Suyarto, R., Diara, I. W., Susila, K. D., Saifulloh, M., Wiyanti, W., Kusmiyarti, T. B., & Sunarta, I. N. (2023). Landslide inventory mapping derived from multispectral imagery by support vector machine (SVM) algorithm. In IOP Conference Series: Earth and Environmental Science, 1190(1), 012012. https://doi.org/10.1088/1755-1315/1190/1/012012

Trigila, A., Iadanza, C., Esposito, C., & Scarascia-Mugnozza, G. (2015). Comparison of Logistic Regression and Random Forests techniques for shallow landslide susceptibility assessment in Giampilieri (NE Sicily, Italy). Geomorphology, 249, 119-136.‏ https://doi.org/10.1016/j.geomorph.2015.06.001

Wang, Y., Fang, Z., Wang, M., Peng, L., & Hong, H. (2020). Comparative study of landslide susceptibility mapping with different recurrent neural networks. Computers & Geosciences, 138, 104445.‏ https://doi.org/10.1016/j.cageo.2020.104445

Yao, J., Qin, S., Qiao, S., Liu, X., Zhang, L., & Chen, J. (2022). Application of a two-step sampling strategy based on deep neural network for landslide susceptibility mapping. Bulletin of Engineering Geology and the Environment, 81(4), 148.‏ https://doi.org/10.1007/s10064-022-02615-0

Yeon, Y. K., Han, J. G., & Ryu, K. H. (2010). Landslide susceptibility mapping in Injae, Korea, using a decision tree. Engineering Geology, 116(3-4), 274-283.‏ https://doi.org/10.1016/j.enggeo.2010.09.009

Yu, H., Pei, W., Zhang, J., & Chen, G. (2023). Landslide susceptibility mapping and driving mechanisms in a vulnerable region based on multiple machine learning models. Remote Sensing, 15(7), 1886.‏ https://doi.org/10.3390/rs15071886

Zevenbergen, L. W., & Thorne, C. R. (1987). Quantitative analysis of land surface topography. Earth Surface Processes and Landforms, 12(1), 47-56. https://doi.org/10.1002/esp.3290120107

Zhao, Z., Liu, Z. Y., & Xu, C. (2021). Slope unit-based landslide susceptibility mapping using certainty factor, support vector machine, random forest, CF-SVM and CF-RF models. Frontiers in Earth Science, 9, 589630.‏ https://doi.org/10.3389/feart.2021.589630