A Comparative Study of TabNet and Classical Machine Learning Models for Landslide Prediction

A Comparative Study of TabNet and Classical Machine Learning Models for Landslide Prediction | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A Comparative Study of TabNet and Classical Machine Learning Models for Landslide Prediction Ali Aalianvari, Shirin Jahanmiri This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7694229/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Landslides are a major geohazard that endanger human life, infrastructure, and ecosystems, highlighting the need for accurate susceptibility mapping to support proactive disaster risk management. Traditional machine learning models such as Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Networks (ANN), and XGBoost have shown utility but often struggle to capture the complex, high-dimensional interactions among heterogeneous geospatial factors. This study introduces an advanced framework based on TabNet , a deep learning architecture optimized for tabular data, which leverages sequential attention and interpretable decision steps to effectively model intricate feature relationships. A synthetic yet realistic dataset of 180 samples and 21 conditioning factors—including slope, elevation, rainfall, land use, lithology, and proximity to faults and rivers—was constructed from geospatial patterns reported in 17 peer-reviewed studies. TabNet was benchmarked against four classical models. Results show that TabNet achieved the highest predictive performance with an accuracy of 87%, AUC-ROC of 0.92, and F1-score of 0.82, outperforming all baseline models. Feature importance analysis identified slope, rainfall intensity, and land cover as the most critical predictors of landslide occurrence. Moreover, TabNet demonstrated strong generalization across diverse synthetic terrains resembling the Himalayas, the Alps, Southeast Asia, and the Zagros Mountains, while maintaining low misclassification rates and competitive training efficiency. These findings highlight TabNet’s robustness, interpretability, and superior predictive capacity, positioning it as a promising tool for geospatial hazard assessment. Future work will focus on integrating temporal rainfall records, high-resolution remote sensing, and real-time seismic data to enhance predictive responsiveness and enable operational deployment in early warning systems. Earth and environmental sciences/Environmental sciences Earth and environmental sciences/Natural hazards Landslide Susceptibility Mapping (LSM) TabNet Deep Learning Environmental Conditioning Factors Feature Selection Hazard Prediction Full Text Additional Declarations No competing interests reported. Supplementary Files landslidedata.xlsx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7694229","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":530532916,"identity":"d95cedcd-5f8e-4a9c-b80b-02955cc13ed7","order_by":0,"name":"Ali Aalianvari","email":"data:image/png;base64,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","orcid":"","institution":"Tarbiat Modares University","correspondingAuthor":true,"prefix":"","firstName":"Ali","middleName":"","lastName":"Aalianvari","suffix":""},{"id":530532917,"identity":"0ee52620-9b9b-44a0-acda-00fe870be375","order_by":1,"name":"Shirin Jahanmiri","email":"","orcid":"","institution":"University of Kashan","correspondingAuthor":false,"prefix":"","firstName":"Shirin","middleName":"","lastName":"Jahanmiri","suffix":""}],"badges":[],"createdAt":"2025-09-23 12:23:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7694229/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7694229/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93825272,"identity":"0c1d18b6-1088-4ba6-85a7-2ea230b1e741","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3688004,"visible":true,"origin":"","legend":"","description":"","filename":"LandslideSusceptibilityPredictionUsingTabNet.docx","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/2f9ba91433ce88de3ddee52d.docx"},{"id":93825256,"identity":"027d9f2c-627f-4db4-8336-1f36a5f4ce78","added_by":"auto","created_at":"2025-10-18 06:09:38","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4955,"visible":true,"origin":"","legend":"","description":"","filename":"b6303e5a63b64b7dbda133791fe4a983.json","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/91b7f168b55b4df78bbc7c5f.json"},{"id":93825259,"identity":"5a1f33dc-a51a-4765-af79-6fded249223f","added_by":"auto","created_at":"2025-10-18 06:09:38","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":71485,"visible":true,"origin":"","legend":"","description":"","filename":"landslidedata.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/80367731c71c8840284825a7.xlsx"},{"id":93825263,"identity":"8a00ad01-e0d4-4536-ae56-f82c28d494d3","added_by":"auto","created_at":"2025-10-18 06:09:38","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":175166,"visible":true,"origin":"","legend":"","description":"","filename":"b6303e5a63b64b7dbda133791fe4a9831enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/a5d33b5383b0ed72c0c2fbf0.xml"},{"id":93825497,"identity":"a79e09d8-db40-4f5d-97bd-c9b5534b3dc8","added_by":"auto","created_at":"2025-10-18 06:17:38","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":219053,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/0e1697c47891fca96d659cce.png"},{"id":93825496,"identity":"4a14d7ad-6e56-4feb-9104-b8c9e2855ac6","added_by":"auto","created_at":"2025-10-18 06:17:38","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39634,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/7fb3130e444cfc59d8947938.png"},{"id":93825258,"identity":"66433525-1ab3-47d1-93ab-74a22343f4a8","added_by":"auto","created_at":"2025-10-18 06:09:38","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31267,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/e5c2ebc7a1535a20ce15799a.png"},{"id":93825498,"identity":"f2b9960b-1661-480b-90f9-eb14f99e7983","added_by":"auto","created_at":"2025-10-18 06:17:38","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":76817,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/9180cc28baaf28ad722d5b4e.png"},{"id":93825968,"identity":"441f5dbd-09e6-49c4-a1d9-c0a9164efe2c","added_by":"auto","created_at":"2025-10-18 06:25:38","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":59752,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage13.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/bddca0d5b5fed632a635735b.png"},{"id":93825264,"identity":"b8656a40-aa5d-48ea-a71f-c16bf8d7dbc1","added_by":"auto","created_at":"2025-10-18 06:09:38","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":89667,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage14.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/23f6a965e34ebd7ecc30b4c8.png"},{"id":93825969,"identity":"2bc60833-aacb-4255-9ac1-6113a2779001","added_by":"auto","created_at":"2025-10-18 06:25:38","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27849,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage15.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/cb31e7eab8cee9a1d1f40416.png"},{"id":93825274,"identity":"67decc3d-b4b2-4f1e-9f8f-0690a7fb8d8a","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29455,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage16.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/cb10fb49f4e95c098e3bb72f.png"},{"id":93825266,"identity":"204cfa91-b1b2-4128-8760-1dbb814495b1","added_by":"auto","created_at":"2025-10-18 06:09:38","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129322,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/eac2da7428149aa48e23f8b8.png"},{"id":93825970,"identity":"4387e49a-d15f-4f38-8503-49a7531c6c54","added_by":"auto","created_at":"2025-10-18 06:25:39","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2093737,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/2c9cf67e95fecd91e2638537.jpeg"},{"id":93825505,"identity":"86be55ee-cb00-4249-a58b-b7a7698c96ff","added_by":"auto","created_at":"2025-10-18 06:17:39","extension":"jpeg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2279053,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/f5f7df9ec28f6fb9365e1937.jpeg"},{"id":93825276,"identity":"399e4676-f1bd-445b-93a8-78f61f2285d4","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"jpeg","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2247168,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/4906d3855d110c356f747efe.jpeg"},{"id":93825275,"identity":"1e0a8dec-5110-42ae-a053-eb8ad4233c10","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"jpeg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":440865,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/8e516b24ab0ac4bf518b7fb1.jpeg"},{"id":93825282,"identity":"78adbf3d-6ede-4c00-a12e-cb5ab895334f","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32445,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/e633c69034eac0bc375d9835.png"},{"id":93825267,"identity":"724a9bc5-0259-4082-9f8c-0491c6eb0311","added_by":"auto","created_at":"2025-10-18 06:09:38","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30786,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/ca4b8445aafce2a2c283ed8a.png"},{"id":93825508,"identity":"8f8cace7-28c0-4942-bee2-526ac779042b","added_by":"auto","created_at":"2025-10-18 06:17:39","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32274,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/f4e622746eb82a4a17b54f85.png"},{"id":93825503,"identity":"8444559d-4d99-496c-816c-2f904a446939","added_by":"auto","created_at":"2025-10-18 06:17:39","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30138,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/062c69e59de777eb16884950.png"},{"id":93825500,"identity":"aa8d35a7-c67e-4914-b41c-b9a62eb8f97a","added_by":"auto","created_at":"2025-10-18 06:17:38","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12468,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/ae80b9c9d7ae357a48fc3300.png"},{"id":93825506,"identity":"f3fc1cf9-6253-4bff-a100-1a221263b9ad","added_by":"auto","created_at":"2025-10-18 06:17:39","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13825,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/84f80d211482f7d991be8b54.png"},{"id":93825270,"identity":"f6c9e4b6-525b-4e6a-90aa-d9825e0a9615","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20265,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/38ed2ed6acd3a8538489b93a.png"},{"id":93825501,"identity":"b9860630-bd11-4444-9409-1ebe01836b45","added_by":"auto","created_at":"2025-10-18 06:17:38","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16438,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage13.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/8e3a3d04ea0c4f8dc0ce8acb.png"},{"id":93825507,"identity":"25956997-9cb8-4503-aa1f-542907328735","added_by":"auto","created_at":"2025-10-18 06:17:39","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22009,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage14.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/415be00601f0c81d792deb33.png"},{"id":93825287,"identity":"6ce3c05a-36af-4227-9cfb-470f65fdf37f","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12327,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage15.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/9ae9f7bfca0d5947fa64931a.png"},{"id":93825281,"identity":"2b69af1b-7c68-486b-8a39-207c20d466f0","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9568,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage16.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/6dcdfffb3213febef32e1547.png"},{"id":93825510,"identity":"183395c9-396c-4e9e-89a7-fea15fdfe991","added_by":"auto","created_at":"2025-10-18 06:17:39","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32045,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/bc4f193d817f5ab9ef47edfd.png"},{"id":93825279,"identity":"262ca72a-373d-455b-a130-f7d42c40b1bc","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":813478,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/bbbb4ea07915352d4610f2b2.png"},{"id":93825285,"identity":"cc170875-7a17-432d-8b6a-53ca249fe7eb","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":770396,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/4e50bb92bbd7f0bda30e9740.png"},{"id":93825280,"identity":"3d9e8267-4d3c-4342-85cb-d60eea39ae36","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":662005,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/de5b56e46c742046eaed9621.png"},{"id":93825971,"identity":"3c9fd720-39d9-497a-9d0d-7c3995f034c7","added_by":"auto","created_at":"2025-10-18 06:25:39","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102307,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/a7fa445bcc6318f2dfdf4fca.png"},{"id":93825511,"identity":"6eb41e71-7e6c-4e72-97ec-2a431c164e85","added_by":"auto","created_at":"2025-10-18 06:17:39","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14257,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/3a295e20397b4c230c16d8f6.png"},{"id":93825284,"identity":"69e78283-1c4a-4419-9eb4-4fd81ae20af1","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14769,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/7f47d4b8c765297823cf06c8.png"},{"id":93825278,"identity":"7fdc12ce-14d7-44e7-95c4-e02c3a50f581","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"png","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14411,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/aa9ce2f5cbe21a393df95ba5.png"},{"id":93825290,"identity":"c402e9e4-b578-467e-9e02-f290012fe324","added_by":"auto","created_at":"2025-10-18 06:09:39","extension":"xml","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":172692,"visible":true,"origin":"","legend":"","description":"","filename":"b6303e5a63b64b7dbda133791fe4a9831structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/bab410b54381343c99409d9b.xml"},{"id":93825509,"identity":"74897f8c-f202-4f82-a233-b1dfc1966579","added_by":"auto","created_at":"2025-10-18 06:17:39","extension":"html","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":192577,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/2560a581183e6ad93bcfed34.html"},{"id":101548949,"identity":"9b0a0b58-fbd1-42e9-a384-cacc776f4a13","added_by":"auto","created_at":"2026-01-31 06:40:08","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2163900,"visible":true,"origin":"","legend":"","description":"","filename":"LandslideSusceptibilityPredictionUsingTabNet.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1_covered_bc278267-6a9e-40bb-b49f-c6665c0064a6.pdf"},{"id":93825255,"identity":"10b20aec-5e2a-41c3-b719-2a3377b9239b","added_by":"auto","created_at":"2025-10-18 06:09:38","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":71485,"visible":true,"origin":"","legend":"","description":"","filename":"landslidedata.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7694229/v1/ebaf198a23b0b43116558a18.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Comparative Study of TabNet and Classical Machine Learning Models for Landslide Prediction","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Landslide Susceptibility Mapping (LSM), TabNet, Deep Learning, Environmental Conditioning Factors, Feature Selection, Hazard Prediction","lastPublishedDoi":"10.21203/rs.3.rs-7694229/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7694229/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLandslides are a major geohazard that endanger human life, infrastructure, and ecosystems, highlighting the need for accurate susceptibility mapping to support proactive disaster risk management. Traditional machine learning models such as Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Networks (ANN), and XGBoost have shown utility but often struggle to capture the complex, high-dimensional interactions among heterogeneous geospatial factors. This study introduces an advanced framework based on \u003cb\u003eTabNet\u003c/b\u003e, a deep learning architecture optimized for tabular data, which leverages sequential attention and interpretable decision steps to effectively model intricate feature relationships. A synthetic yet realistic dataset of 180 samples and 21 conditioning factors\u0026mdash;including slope, elevation, rainfall, land use, lithology, and proximity to faults and rivers\u0026mdash;was constructed from geospatial patterns reported in 17 peer-reviewed studies. TabNet was benchmarked against four classical models. Results show that TabNet achieved the highest predictive performance with an accuracy of 87%, AUC-ROC of 0.92, and F1-score of 0.82, outperforming all baseline models. Feature importance analysis identified slope, rainfall intensity, and land cover as the most critical predictors of landslide occurrence. Moreover, TabNet demonstrated strong generalization across diverse synthetic terrains resembling the Himalayas, the Alps, Southeast Asia, and the Zagros Mountains, while maintaining low misclassification rates and competitive training efficiency. These findings highlight TabNet\u0026rsquo;s robustness, interpretability, and superior predictive capacity, positioning it as a promising tool for geospatial hazard assessment. Future work will focus on integrating temporal rainfall records, high-resolution remote sensing, and real-time seismic data to enhance predictive responsiveness and enable operational deployment in early warning systems.\u003c/p\u003e","manuscriptTitle":"A Comparative Study of TabNet and Classical Machine Learning Models for Landslide Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-18 06:09:33","doi":"10.21203/rs.3.rs-7694229/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3a9b5da5-10f3-4b4a-80a9-15406b0cf55f","owner":[],"postedDate":"October 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":56400519,"name":"Earth and environmental sciences/Environmental sciences"},{"id":56400520,"name":"Earth and environmental sciences/Natural hazards"}],"tags":[],"updatedAt":"2026-01-31T06:39:30+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-18 06:09:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7694229","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7694229","identity":"rs-7694229","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}