International Journal of Science, Engineering and TechnologyAuthors: Mrs. Aruna C, S. Sumalatha
Abstract: In this day and age, cyberattacks have become increasingly sophisticated such that traditional authentication mechanisms like passwords can no longer secure sensitive information. In this paper, DeepSecAuth is proposed, an advanced authentication scheme that integrates keystroke dynamics behavioral biometrics with deepFace-based face recognition technology with the help of a honeypot deception technique. DeepSecAuth uses the one-class support vector machine algorithm to learn the unique typing pattern of the user at the time of registration, achieving 92% verification accuracy while the DeepFace achieves 95% recognition accuracy as the second level of authentication. The experimental results show that the integrated system has achieved 97% overall authentication accuracy with FRR of 1.2% and FAR of 1.6%, which is far better than the unimodal authentication schemes. The honeypot technique ensures that the credentials theft and spoofing attack are mitigated.
DOI: http://doi.org/10.5281/zenodo.21675751
