This study examined the potential use of keystroke dynamics to create keyprints (typing fingerprints) to authenticate individuals in online assessment situations. The implications of this study are best understood in terms of the keystroke behavioral biometric. While previous studies considered the degree to which keystroke typing patterns are unique, this study was set up to determine how well keyprints are able to identify individuals when typing under various treatment conditions (copy typing, free typing, and typing with mild or moderate impediments). While authentication can be difficult when attempting to correctly identify individual users, the results of this study indicate that keyprints can be a solid indicator of negative cases (i.e., flagging situations where a typing sample is likely not the correct individual). As anticipated, typing with a temporary impediment does diminish the algorithms' ability to identify students. This is also the case when user samples are typed under conditions different from those in which the keyprint baseline signature was captured (i.e., copy versus free typing). The ability to identify individuals is also challenging when using small comparison samples. However, the ability of the system to identify negative cases functions fairly well in each instance.
| Identifer | oai:union.ndltd.org:BGMYU2/oai:scholarsarchive.byu.edu:etd-7690 |
| Date | 01 February 2018 |
| Creators | Young, Jay Richards |
| Publisher | BYU ScholarsArchive |
| Source Sets | Brigham Young University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | All Theses and Dissertations |
| Rights | http://lib.byu.edu/about/copyright/ |
Page generated in 0.0013 seconds