Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68022
Volume:
10 |
10 |
Oct 2026
Pages:
1-6
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department of Computer Application
BBD UNIVERSITY, LUCKNOW
Department of Computer Application
BBD UNIVERSITY, LUCKNOW
Department of Computer Application
BBD UNIVERSITY, LUCKNOW
Department of Computer Application
BBD University, Lucknow - 226 028, India
Cloud platforms keep valuable data behind a single login, so stolen or guessed credentials remain one of the easiest ways in. Static multi-factor authentication (MFA) closes much of that gap, but it asks every user for an extra factor on every login, and that friction tempts organisations to weaken their policies. This paper presents RA-MFA, a risk-adaptive MFA framework for cloud applications. After the password check, a lightweight logistic-regression model scores each login attempt from eight contextual signals such as IP reputation, travel velocity, device novelty and behavioural deviation. The score selects one of three responses: allow with a device token, a push/TOTP challenge, or a strong challenge (TOTP plus biometric or a FIDO2 key). We evaluate the framework on a synthetic dataset of 20,000 login events that contains 1,577 account-takeover attempts, and compare it with password-only, static TOTP and rule-based adaptive baselines. On the held-out test set RA-MFA blocked 95.8% of attacks, which is comparable to static MFA (94.7%), while it challenged only 22.9% of legitimate logins and reduced the average added login time from 11.0 s to 2.2 s. We also discuss the limits of simulated evaluation and describe how the framework can be validated on real traffic.
KEYWORDS
This paper presented RA-MFA, a risk-adaptive multi-factor authentication framework that uses a transparent logistic-regression score to decide how much proof a cloud login requires. In a controlled simulation with 20,000 synthetic login events, it matched the protection of static TOTP (4.2% versus 5.3% attack success, a difference within sampling error) while challenging fewer than a quarter of genuine logins and cutting added login time by about 80%. It also clearly outperformed a hand-weighted rule-based baseline, which let 15.4% of attacks through. Future work will proceed along four lines: validating the framework on anonymised real login logs in shadow mode; replacing logistic regression with gradient-boosted or sequence models and comparing their interpretability; adding continuous authentication during the session; and studying the privacy cost of behavioural features, possibly with on-device scoring.
[1] M. P. D. Prakash and A. Singh, “Maximizing cloud security: Synergistic approaches to multi-factor authentication deployment,” in Proc. Int. Conf. Recent Advancement and Modernization in Sustainable Intelligent Technologies & Applications (RAMSITA-2025), Adv. Intell. Syst. Res., vol. 192, 2025, pp. 945–955, doi: 10.2991/978-94-6463-716-8_70. [2] R. R. Pell Reddy, “Empowering cloud security: Pioneering an interactive multi-factor authentication framework for cloud user verification,” Int. J. Comput. Trends Technol., vol. 72, no. 6, pp. 168–178, Jun. 2024, doi: 10.14445/22312803/IJCTT-V72I6P122. [3] A. H. M. Aburbeian and M. Fernández-Veiga, “Secure internet financial transactions: A framework integrating multi-factor authentication and machine learning,” AI, vol. 5, no. 1, pp. 177–194, 2024, doi: 10.3390/ai5010010. [4] R. O. Okeke and S. O. Orimadike, “Enhanced cloud computing security using application-based multi-factor authentication (MFA) for communication systems,” Eur. J. Electr. Eng. Comput. Sci., vol. 8, no. 2, pp. 1–8, 2024, doi: 10.24018/ejece.2024.8.2.593. [5] Y. Han, H. Guo, J. Liu, B. B. Ehui, Y. Wu, and S. Li, “An enhanced multifactor authentication and key agreement protocol in industrial Internet of Things,” IEEE Internet Things J., vol. 11, pp. 16243–16254, 2024, doi: 10.1109/JIOT.2024.3355228. [6] S. Shukla and S. J. Patel, “A design of provably secure multi-factor ECC-based authentication protocol in multi-server cloud architecture,” Cluster Comput., vol. 27, pp. 1559–1580, 2024. [7] R. Konwar, D. Jha, R. Agrawal, R. Purkayastha, and I. Banerjee, “A two-factor authentication mechanism using a novel OTP generation algorithm for cloud applications,” in Proc. 14th Int. Conf. Cloud Comput., Data Sci. Eng. (Confluence), 2024, pp. 245–250, doi: 10.1109/Confluence60223.2024.10463309.
Publication Date
10/15/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
3
Total views since publication
PDF Downloads
11
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...