AI-Driven Behavioral Biometrics and Anomaly Detection for Account Takeover and Identity Theft Prevention
A Comprehensive Survey
الكلمات المفتاحية:
artificial intelligence، behavioral biometrics، anomaly detection، fraud detection، cybersecurity، machine learningالملخص
The rapid growth of digital services and online transactions has significantly increased the risk of cyber threats, particularly account takeover (ATO) attacks and identity theft. Traditional authentication mechanisms, such as passwords and one-time passcodes, are increasingly vulnerable to sophisticated attack techniques including phishing, credential stuffing, and social engineering. To address these challenges, behavioral biometrics and anomaly detection have emerged as promising approaches for continuous and adaptive user authentication. Behavioral biometrics leverage unique patterns of user behavior, such as keystroke dynamics, mouse movements, touchscreen interactions, gait, and device usage patterns, to establish and verify user identity. When combined with Artificial Intelligence (AI) and Machine Learning (ML) techniques, these systems can effectively identify abnormal activities and potential security breaches in real time.
This comprehensive survey reviews recent advancements in AI-driven behavioral biometrics and anomaly detection methods for account takeover and identity theft prevention. It examines various behavioral biometric modalities, data collection techniques, feature extraction methods, and AI-based classification and anomaly detection algorithms, including supervised, unsupervised, and deep learning approaches. The survey further analyzes publicly available datasets, performance evaluation metrics, deployment challenges, privacy concerns, and security vulnerabilities associated with behavioral biometric systems. In addition, it highlights emerging trends such as continuous authentication, explainable AI, federated learning, and multimodal behavioral analysis.
التنزيلات
المراجع
Baig, A. F., Eskeland, S., & Yang, B. (2024). Novel and efficient privacy-preserving continuous authentication. Cryptography, 8(1), 3.
Dehling, F., et al. (2024). Internet users' willingness to disclose biometric data for continuous online account protection. Proceedings on Privacy Enhancing Technologies.
Li, J., et al. (2024). MBBFAuth: Multimodal behavioral biometrics fusion for continuous authentication on non-portable devices. IEEE Transactions on Information Forensics and Security.
Murthy, S. D. (2026). Identity theft detection at data ingestion using AI: An explainable anomaly detection approach. SSRN.
Obrik-Uloho, E. (2025). Elevating continuous verification through advanced behavioral analytics. Journal of Engineering Research and Reports.
Olabiyi, W. (2025). Behavioral biometrics and continuous authentication for financial fraud prevention.
Yusuf, I., et al. (2024). AI-powered behavioral biometrics for continuous authentication in cybersecurity. International Journal of Information Security.
Zhang, X., et al. (2024). MotionID: Towards practical behavioral biometrics-based implicit user authentication on smartphones. Pervasive and Mobile Computing, 101.
التنزيلات
-
Download PDF (الإنجليزية)
الملخص Views: 0,
Download PDF (الإنجليزية): 0
منشور
إصدار
القسم
الرخصة
الحقوق الفكرية (c) 2026 International Journal of Innovative Scientific Research

هذا العمل مرخص بموجب Creative Commons Attribution-ShareAlike 4.0 International License.












