Explainable AI-Driven Hybrid Feature Fusion for Robust Android Malware Detection

Explainable AI-Driven Hybrid Feature Fusion for Robust Android Malware Detection | 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 Research Article Explainable AI-Driven Hybrid Feature Fusion for Robust Android Malware Detection Mohd Fozla Rabby, Teresa Jency Bala This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7743592/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 Android malware is growing rapidly, creating an urgent need for powerful and interpretable detection methods. Mobile security needs better ways to find and understand threats. The goal of this study is to create a strong, clear framework for finding Android malware. In this paper we have presented a hybrid feature fusion methodology that integrates static metadata (permissions, intents), dynamic API requests (exceeding 23,000 behavioral indicators), and DEX image-based features derived from a convolutional neural network. The approach uses a dataset of 74,268 APK samples from a public repository. Of these, 18,440 are benign and 55,828 are malicious. The dataset is split into 70% for training, 10% for validation, and 20% for testing. The experimental results show that we got an optimal outcome of 100% accuracy on the test set, suggesting a strong effectiveness of the hybrid approach. The use of explainable AI approaches helps to find the most important factors that affect classification, like questionable permissions and API requests related to the network. These techniques make it easier for nontechnical users to understand without the need to know the underlying complexity. This work is new because it combines multi-modal characteristics with interpretable AI to create a scalable, clear solution for finding Android malware in the real world. Android malware detection explainable AI SHAP static analysis dynamic analysis DEX image features XGBoost mobile security Full Text Additional Declarations No competing interests reported. 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-7743592","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":522366763,"identity":"ed245ec6-583e-4a8b-95e6-6f56858b2eb1","order_by":0,"name":"Mohd Fozla Rabby","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIie2RsYrCQBCGxxPWZiDtSCS+giHgCb5M0phGra0kNqbxAZTzYQILsVn72AlpVQI2FiKOVlftaie4X7W7zMf8MwtgsXwkKAAmhO1Gwpfs+SQAjYrqef4ie0epzScBFOGLym+6zUsUFM1WZVlVSoKTZmJ/0CgtNY4DRIpSd9BtLgsJpMKGv9YoBMOui8Rd/oaijpUEKEC4umDkHFjpUJTsVP18ZaVtVOjRJSQeH8EFDtYxK6fYX2fES+ZZFipGX0WpfhZnlO+Ptyl/pSyrS973vI3MtRt78PM/Bp9riUHgkouxxGKxWL6aO2rzSxn5QAYQAAAAAElFTkSuQmCC","orcid":"","institution":"Khulna University","correspondingAuthor":true,"prefix":"","firstName":"Mohd","middleName":"Fozla","lastName":"Rabby","suffix":""},{"id":522366764,"identity":"dc821dcd-bfe5-42d5-95a3-1970efbcfcfb","order_by":1,"name":"Teresa Jency Bala","email":"","orcid":"","institution":"Khulna University","correspondingAuthor":false,"prefix":"","firstName":"Teresa","middleName":"Jency","lastName":"Bala","suffix":""}],"badges":[],"createdAt":"2025-09-29 15:23:49","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7743592/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7743592/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92489094,"identity":"f93a488c-8be2-4aa3-9e51-b2b729a71633","added_by":"auto","created_at":"2025-09-30 09:16:06","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4034,"visible":true,"origin":"","legend":"","description":"","filename":"6a6d81a0239a47f5b6ad074e6fa836d9.json","url":"https://assets-eu.researchsquare.com/files/rs-7743592/v1/c4daf0b578cfab16735819da.json"},{"id":92489098,"identity":"37f4ef16-09c3-4f5e-b5a9-463b63ad28dd","added_by":"auto","created_at":"2025-09-30 09:16:11","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1015774,"visible":true,"origin":"","legend":"","description":"","filename":"ExplainableAIDrivenHybridFeatureFusionforRobustAndroidMalwareDetection.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7743592/v1_covered_4f1abc49-fd00-4fa6-94a9-20d2e538ed75.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Explainable AI-Driven Hybrid Feature Fusion for Robust Android Malware Detection","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"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":"Android malware detection, explainable AI, SHAP, static analysis, dynamic analysis, DEX image features, XGBoost, mobile security","lastPublishedDoi":"10.21203/rs.3.rs-7743592/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7743592/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAndroid malware is growing rapidly, creating an urgent need for powerful and interpretable detection methods. Mobile security needs better ways to find and understand threats. The goal of this study is to create a strong, clear framework for finding Android malware. In this paper we have presented a hybrid feature fusion methodology that integrates static metadata (permissions, intents), dynamic API requests (exceeding 23,000 behavioral indicators), and DEX image-based features derived from a convolutional neural network. The approach uses a dataset of 74,268 APK samples from a public repository. Of these, 18,440 are benign and 55,828 are malicious. The dataset is split into 70% for training, 10% for validation, and 20% for testing. The experimental results show that we got an optimal outcome of 100% accuracy on the test set, suggesting a strong effectiveness of the hybrid approach. The use of explainable AI approaches helps to find the most important factors that affect classification, like questionable permissions and API requests related to the network. These techniques make it easier for nontechnical users to understand without the need to know the underlying complexity. This work is new because it combines multi-modal characteristics with interpretable AI to create a scalable, clear solution for finding Android malware in the real world.\u003c/p\u003e","manuscriptTitle":"Explainable AI-Driven Hybrid Feature Fusion for Robust Android Malware Detection","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-30 09:15:58","doi":"10.21203/rs.3.rs-7743592/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":"414c683c-59c6-4b2e-847e-e866da927938","owner":[],"postedDate":"September 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-06T12:23:20+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-30 09:15:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7743592","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7743592","identity":"rs-7743592","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}