ATK
New Delhi [India], November 22: Researchers have developed a novel method for distinguishing between multiple users accessing a single application using keystroke dynamics, a technique that analyzes an individual's unique typing patterns. This innovation holds significant promise for enhancing security in shared computing environments, where multiple users have authorized access to common devices and accounts. The paper titled "A novel non-linear transformation based multi user identification algorithm for fixed text keystroke behavioral dynamics" was published in IEEE Transactions on Biometrics, Behavior, and Identity Science and available at arxiv: http://arxiv.org/abs/2210.02505v1
Traditional login methods often become obsolete in these shared settings, leaving systems vulnerable to unauthorized access. To address this challenge, the researchers devised an algorithm that utilizes the quantile transform and localization techniques to effectively classify and identify users. The algorithm, known as ordinal Unfolding-based Localization (UNLOC), operates solely on ordinal data derived from distance proxies, enabling it to distinguish between users based on their distinct typing patterns.
Extensive testing using benchmark keystroke datasets has demonstrated the superior performance of UNLOC compared to existing methods. This development has the potential to revolutionize security practices in shared computing environments, safeguarding sensitive information and preventing unauthorized access.
Key Highlights:
* A new technique has been developed to uniquely classify and identify multiple users accessing a single application using keystroke dynamics.
* This method is particularly useful in shared computing environments where traditional login methods are bypassed.
* The algorithm, known as ordinal Unfolding-based Localization (UNLOC), utilizes the quantile transform and localization techniques to effectively distinguish between users based on their unique typing patterns.
* UNLOC has demonstrated superior performance compared to existing methods in benchmark keystroke datasets.
Related papers: 1. A novel distance-based algorithm for multi-user classification in keystroke dynamics, IEEE: 10.1109/IEEECONF51394.2020.9443407
2. A nonlinear feature transformation-based multi-user classification algorithm for keystroke dynamics, IEEE: 10.1109/IEEECONF53345.2021.9723223
(ADVERTORIAL DISCLAIMER: The above press release has been provided by ATK. ANI will not be responsible in any way for the content of the same)