Hybrid Golden Jackal-Sea Lion and Sea Horse Optimization Algorithm for Improved Keystroke Dynamics User Authentication

In the present-day digital environment, the safeguarding of our online identities has emerged as a matter of utmost significance. Traditionally, individuals predominantly depended on passwords (Single Factor Authentication) as a mechanism for safeguarding their accounts, akin to fortifying a door with a rudimentary latch, thereby making it susceptible to determined intruders. However, the proposed approach, hGJSLnO, employs a hybrid Golden Jackal - Sea Lion Optimization for feature selection and the Sea Horse Optimization algorithm (SHO) for enhanced accuracy, providing keystroke dynamics as a form of Multi-Factor Authentication (MFA). This emerges as a sophisticated security paradigm for online accounts, introducing additional verification layers such as mobile-generated codes or biometric scans like fingerprints. These measures significantly improve security by making unauthorized access difficult, even in cases of password compromise. This paper presents an efficient approach (hGJ-SLnO) utilizing hybrid swarm optimization techniques to separate legitimate users from impostors based on keystroke dynamics. Our proposed methodology is rigorously evaluated using the Carnegie Mellon University (CMU) keystroke benchmark dataset, achieving an average accuracy of 94.3% and an average equal error rate (EER) of 0.034. These findings exceed previous advancements, signaling the efficacy and promise of keystroke dynamics in strengthening online security.