Keystroke dynamics—the analysis of typing rhythms to verify the identity of the person producing the keystrokes—has the potential to improve password-based authentication system by introducing a biometric component that requires no extra hardware cost and provides non-intrusive verification. Various algorithms have been proposed in the literature, and numerous public datasets have attempted to build a body of training data. Despite the availability of public datasets, very few studies have looked at the performance of algorithms across datasets. This study introduces a new algorithm to this domain, a deep neural network, and applies it across two keystroke dynamic datasets. Using the DSL2009 benchmark dataset, we were able to achieve results equivalent to those of existing algorithms, while greatly reducing network size and training time. Using the GREYC dataset and applying the same approach, we were able to achieve an EER of 5.6%, an 18% reduction over the best published methods.
|School:||University of Connecticut|
|School Location:||United States -- Connecticut|
|Source:||MAI 82/6(E), Masters Abstracts International|
|Subjects:||Computer science, Artificial intelligence|
|Keywords:||Keystroke dynamics, Password-based Authentication|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be