IdenTransformer: A Foundation Model Architecture for Robust Digital Identity Verification
Keywords:
Behavioral Biometrics, Digital Security, Foundation Models, Fraud Detection, Identity Verification, Machine Learning, Transaction AnalysisAbstract
This research introduces a novel architecture for digital identity verification by leveraging the representational capabilities of large-scale foundation models. Traditional discriminative approaches for fraud detection often struggle to generalize across heterogeneous identity signals and evolving adversarial behaviors. The proposed framework, IdenTransformer, employs a multimodal transformer architecture that integrates biometric embeddings, behavioral telemetry, and relational metadata through cross-attention mechanisms within a unified latent representation space. Central to the approach is a parameter-efficient fine-tuning strategy using Identity-Adapters, which enables effective adaptation of large pre-trained models to high-cardinality identity verification tasks without requiring full model retraining. The framework further incorporates contrastive representation learning to enforce identity consistency across modalities while preserving sensitivity to anomalous behavioral patterns indicative of synthetic identity creation and coordinated fraud activity. Empirical evaluation on large-scale identity transaction datasets demonstrates significant improvements in fraud detection performance, including a 15% increase in Area Under the Precision–Recall Curve (AUPRC) and a 12% reduction in False Discovery Rate compared with strong ensemble-based baselines. The results demonstrate that foundation-model architectures provide a scalable and robust approach for detecting emerging identity fraud patterns in complex digital ecosystems.
Downloads
References
P. P. Arcot, G. Sayed, B. Parekh, J. V. Balasubramanian, and V. N. Sudheer, “The interplay of ethics, culture, and society in the age of finance digital transformation,” J. Southwest Jiaotong Univ., vol. 59, no. 2, pp. 139–163, 2024.
Ö. Aslan, S. S. Aktuğ, M. Ozkan-Okay, A. A. Yilmaz, and E. Akin, “A comprehensive review of cyber security vulnerabilities, threats, attacks, and solutions,” Electronics, vol. 12, no. 6, p. 1333, 2023.
V. M. Patel et al., “Robust deep learning for computer vision,” Proc. IEEE, 2022.
V. Davidavičienė, J. Raudeliūnienė, M. Tvaronavičienė, and J. Kaušinis, “The importance of security aspects in consumer preferences in electronic environment,” J. Secur. Sustain. Issues, vol. 8, no. 3, 2019.
S. Kumar and S. Prasanna, “Heterogeneous ensemble learning for robust adversarial pattern recognition in digital ecosystems,” J. Comput. Anal. Appl., vol. 27, no. 5, pp. 18–28, 2019.
V. Štruc et al., “Deep learning based face recognition: A survey,” IEEE Access, 2022.
S. T. Boppiniti, “Big data meets machine learning: Strategies for efficient data processing and analysis in large datasets,” Int. J. Creative Res. Comput. Technol. Des., vol. 2, no. 2, 2020.
S. Kumar, S. Prasanna, and X. Ruan, “A unified hybrid machine learning architecture for robust identity anomaly detection in large-scale digital ecosystems,” J. Electr. Syst., vol. 14, no. 1, pp. 160–173, 2018.
R. Wang and W. Ji, “Computational intelligence for information security: A survey,” IEEE Trans. Emerg. Top. Comput. Intell., vol. 4, no. 5, pp. 616–629, 2020.
E. Ellavarason, R. Guest, F. Deravi, R. Sanchez-Riello, and B. Corsetti, “Touch-dynamics based behavioural biometrics on mobile devices—a review from a usability and performance perspective,” ACM Comput. Surv., vol. 53, no. 6, pp. 1–36, 2020.
B. Lebichot, T. Verhelst, Y. A. Le Borgne, L. He-Guelton, F. Oble, and G. Bontempi, “Transfer learning strategies for credit card fraud detection,” IEEE Access, vol. 9, pp. 114754–114766, 2021.
E. Ileberi, Y. Sun, and Z. Wang, “A machine learning based credit card fraud detection using the GA algorithm for feature selection,” J. Big Data, vol. 9, no. 1, p. 24, 2022.
P. Kasprowski, Z. Borowska, and K. Harezlak, “Biometric identification based on keystroke dynamics,” Sensors, vol. 22, no. 9, p. 3158, 2022.
I. Stylios, S. Kokolakis, O. Thanou, and S. Chatzis, “Behavioral biometrics and continuous user authentication on mobile devices: A survey,” Inf. Fusion, vol. 66, pp. 76–99, 2021.
S. P. Maniraj, A. Saini, S. Ahmed, and S. Sarkar, “Credit card fraud detection using machine learning and data science,” Int. J. Eng. Res., vol. 8, no. 9, pp. 110–115, 2019.
A. Kaleel and Z. Polkowski, “Credit card fraud detection and identification using machine learning techniques,” Wasit J. Comput. Math. Sci., vol. 2, no. 4, pp. 159–165, 2023.
P. Raghavan and N. El Gayar, “Fraud detection using machine learning and deep learning,” in Proc. Int. Conf. Comput. Intell. Knowl. Econ. (ICCIKE), Dec. 2019, pp. 334–339.
R. K. Chenoori and R. Kavuri, “Online transaction fraud detection using efficient dimensionality reduction and machine learning techniques,” Revue d’Intelligence Artificielle, vol. 36, no. 4, p. 621, 2022.
S. Khatib, “The application of machine learning models in fraud detection and prevention across digital banking channels and payment platforms,” Int. J. Adv. Comput. Methodol. Emerg. Technol., vol. 14, no. 9, pp. 1–7, 2024.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
