DEVELOPMENT OF A MOBILE ANTI-FRAUD MODEL BASED ON SESSION AND TRANSACTION ANALYSIS
Keywords:
anti-fraud, mobile banking, session analysis, transaction monitoring, behavioural biometrics, device fingerprinting, account takeover, authorised push payment fraud, remote access trojan, class imbalance, cost-sensitive learning, concept drift.Abstract
This article addresses the detection of fraud in mobile banking at the point at which a payment is authorised. The prevailing practice evaluates the payment order in isolation, by rules applied to its amount, time and beneficiary. That practice fails against the two forms of attack that now dominate: remote control of the customer device, in which the payment originates from the genuine device and the genuine credentials, and social engineering, in which the customer authorises the payment himself. In both the transaction is formally legitimate and the evidence of fraud lies not in the order but in the session that produced it. A four-layer reference model is constructed in which device, behavioural and transactional signals are assembled into a session-context feature vector, a risk score is computed before authorisation, and the score is mapped to a graduated decision policy whose thresholds are set from the cost of each class of error rather than from a single accuracy figure. A taxonomy of session signals is given with the attack class each signal detects; the construction of the model under extreme class imbalance is described; and an evaluation methodology is stated in which recall at a fixed review capacity, rather than accuracy or the area under the ROC curve, is the operative metric. Adversarial adaptation, concept drift and the requirements of the legislation of the Republic of Uzbekistan are examined.
References
1.Annual Fraud Report 2026: the definitive overview of payment industry fraud in 2025. – London: UK Finance, 2026.
2.UK payment fraud losses topped £1.28bn in 2025, new UK Finance report finds. – URL: https://www.fintechfutures.com/financial-crime-fraud-prevention/uk-payment-fraud-losses-topped-1-28bn-in-2025-new-uk-finance-report-finds (accessed 4 November 2026).
3.UK Finance Fraud Report 2026: payment fraud losses reach £1.28 billion. – URL: https://www.neopay.co.uk/uk-finance-fraud-report-2026-payment-fraud-losses-reach-1-28-billion/ (accessed 4 November 2026).
4.RemControl Banking Trojan Gives Attackers Remote Control of Android Devices. – Infosecurity Magazine, 2026. – URL: https://www.infosecurity-magazine.com/news/banking-trojan-remote-control/ (accessed 4 November 2026).
5.BankBot YNRK Trojan Reinforces Mobile Banking Threats. – Zimperium Mobile Threat Watch, 2026. – URL: https://zimperium.com/blog/mobile-threat-watch/new-android/bankbot-ynrk-trojan-reinforces-mobile-banking-threats (accessed 4 November 2026).
6.Anatsa Banking Trojan Targeting Android Devices: advisory. – Abuja: Nigeria Computer Emergency Response Team, 2026. – URL: https://cert.gov.ng/advisories/anatsa-banking-trojan-targeting-android-devices (accessed 4 November 2026).
7.Preventing Fraud on Mobile Devices Using Behavioral Biometrics: white paper. – Tel Aviv: BioCatch, 2025. – URL: https://www.biocatch.com/resources/white-paper/prevent-fraud-mobile-devices-using-biometric-authentication (accessed 4 November 2026).
8.Top LATAM bank reduces social engineering fraud targeting mobile users with behavioral biometrics: case study. – BioCatch, 2025. – URL: https://www.casestudies.com/company/biocatch/case-study/top-latam-bank-reduces-social-engineering-fraud-targeting-mobile-users-by-67-with-behavioral-biometrics (accessed 4 November 2026).
9.Real-Time Fraud Detection Using AI-Powered Behavioral Biometrics. – GeekyAnts, 2025. – URL: https://geekyants.com/blog/real-time-fraud-detection-using-ai-powered-behavioral-biometrics (accessed 4 November 2026).
10.Dal Pozzolo A., Boracchi G., Caelen O., Alippi C., Bontempi G. Credit Card Fraud Detection: A Realistic Modeling and a Novel Learning Strategy // IEEE Transactions on Neural Networks and Learning Systems. – 2018. – Vol. 29, No. 8. – P. 3784–3797.
11.Dal Pozzolo A., Caelen O., Johnson R.A., Bontempi G. Calibrating Probability with Undersampling for Unbalanced Classification // Proceedings of the IEEE Symposium Series on Computational Intelligence. – 2015. – P. 159–166.
12.Bhattacharyya S., Jha S., Tharakunnel K., Westland J.C. Data mining for credit card fraud: A comparative study // Decision Support Systems. – 2011. – Vol. 50, No. 3. – P. 602–613.
13.Chen T., Guestrin C. XGBoost: A Scalable Tree Boosting System // Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. – 2016. – P. 785–794.
14.Ke G., Meng Q., Finley T. et al. LightGBM: A Highly Efficient Gradient Boosting Decision Tree // Advances in NeurIPS. – 2017. – P. 3146–3154.
15.Prokhorenkova L., Gusev G., Vorobev A. et al. CatBoost: unbiased boosting with categorical features // Advances in NeurIPS. – 2018. – P. 6638–6648.
16.Chawla N.V., Bowyer K.W., Hall L.O., Kegelmeyer W.P. SMOTE: Synthetic Minority Over-sampling Technique // Journal of Artificial Intelligence Research. – 2002. – Vol. 16. – P. 321–357.
17.Elkan C. The Foundations of Cost-Sensitive Learning // Proceedings of IJCAI. – 2001. – P. 973–978.
18.Saito T., Rehmsmeier M. The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets // PLoS ONE. – 2015. – Vol. 10, No. 3. – e0118432.
19.Gama J., Žliobaitė I., Bifet A., Pechenizkiy M., Bouchachia A. A survey on concept drift adaptation // ACM Computing Surveys. – 2014. – Vol. 46, No. 4. – Article 44.
20.Frank M., Biedert R., Ma E., Martinovic I., Song D. Touchalytics: On the Applicability of Touchscreen Input as a Behavioral Biometric for Continuous Authentication // IEEE Transactions on Information Forensics and Security. – 2013. – Vol. 8, No. 1. – P. 136–148.
21.Sitová Z., Šeděnka J., Yang Q. et al. HMOG: New Behavioral Biometric Features for Continuous Authentication of Smartphone Users // IEEE Transactions on Information Forensics and Security. – 2016. – Vol. 11, No. 5. – P. 877–892.
22.Lundberg S.M., Lee S.-I. A Unified Approach to Interpreting Model Predictions // Advances in NeurIPS. – 2017. – P. 4765–4774.
23.Weber M., Domeniconi G., Chen J. et al. Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics // arXiv:1908.02591. – 2019.
24.Guidelines on the security measures for operational and security risks of payment services under PSD2, EBA/GL/2017/17. – Paris: European Banking Authority, 2017.
25.Payments Fraud Report 2025. – Frankfurt: European Banking Authority and European Central Bank, 2025.
26.Digital Identity Guidelines: NIST Special Publication 800-63-3. – Gaithersburg: National Institute of Standards and Technology, 2017.
27.OWASP Mobile Application Security Verification Standard, version 2.0. – Open Worldwide Application Security Project, 2024.
28.On Personal Data: Law of the Republic of Uzbekistan of 2 July 2019, No. ZRU-547 (as amended). – URL: https://lex.uz/ru/docs/4396428
29.On Cybersecurity: Law of the Republic of Uzbekistan of 15 April 2022, No. ZRU-764. – National Database of Legislation of the Republic of Uzbekistan.
30.On Payments and the Payment System: Law of the Republic of Uzbekistan of 1 November 2019, No. ZRU-578 (as amended). – URL: https://lex.uz/docs/4581969
31.On Informatisation: Law of the Republic of Uzbekistan of 11 December 2003, No. 560-II (as amended). – URL: https://lex.uz/ru/docs/82956
32.On the Strategy of Development of New Uzbekistan for 2022–2026: Decree of the President of the Republic of Uzbekistan of 28 January 2022, No. UP-60.