An enhanced multifactor authentication using decision-tree and Logistics-Regression machine learning algorithm

This paper investigates multifactor authentication algorithm which will influence the development of resilient MFA authentication to effectively protect against MFA fatigue attacks. An enhanced MFA algorithm will significantly improve the attack detection abilities which will reinforce the preservation of information confidentiality and integrity. This will result into an improved application security posture. The current MFA Algorithms strengthens the authentication layer by introducing a second authentication factor that verifies and validates the user before access is granted. However, while the current MFA improves authentication it does not address the issue of unsolicited malicious login which are aimed at circumventing MFA Authentication layer without introducing latency between the first (username) and the second authentication factor (biometric). This Study aims to develop an enhanced MFA algorithm that detect and prevent MFA fatigue attacks using both Decision-Tree and Logistics-Regression machine learning algorithm. Our proposed algorithm will be called Decision Tree – Logistics Regression (DTLR). The combined algorithm will improve both the detection and prediction efficacy, while simplifying the MFA authentication factors by only using username and password with one biological factor or MFA mobile application instead of all five authentication factors. Additionally, the proposed algorithm will reduce the latency between the first and second authentication factor. The research methodology will use simulation method to test detection efficacy and performance of DT-LR MFA algorithm. The results will be compared with the existing MFA algorithms to confirm its efficacy level and the superiority of the proposed algorithm. The simulation result will highlight scientific evidence of DT-LR MFA algorithm’s detection efficacy and its effectiveness in detection MFA fatigues attacks. The discoveries of this research work will significantly contribute towards the field of cybersecurity by addressing MFA security challenges, especially in the addressing the challenges of MFA fatigue attacks.