W&M Featured Events
This calendar presented by
William & Mary
[PAST EVENT] Nicolas Van Balen - Dissertation Defense - Computer Science
December 19, 2014
9am - 10:45am
Abstract:
Along the security chain of Internet communications, human users are the weakest link and are vulnerable to Internet fraud and identity theft. User authentication is a critical process of identifying a user and verifying that the user is allowed to access some restricted service and information. An effective user authentication mechanism provides a powerful guard against unauthorized access to systems and data.
In this dissertation proposal, we first propose a graphical password scheme called GridMap, in which maps are used to increase the memorability of a password for users. Moreover, GridMap employs a keyboard based input system for an increased resistance to malware and shoulder surfing attacks. A user study shows that GridMap works well in domains in which a user logs in on a regular basis, and provides a memorability benefit if the chosen map has a personal significance to the user.
To assure a user's actual identity, we propose to use mouse dynamics to determine a user's gender. In this study, mouse movement data was gathered from 96 participants and was used to identify which metrics can provide the largest differences in gender. Using the chosen metrics, we build a classifier for accurate gender identification. In our future work, we will use a similar method to determine a user's age based on mouse dynamics. A larger set of users will be used in this study in order to identify the metrics most useful for age classification, and a model will be built to attempt to determine if a user is 18 years of age or not.
Bio:
Nicolas Van Balen is a fifth year Ph.D. student at the College of William and Mary under the supervision of Dr. Wang. His research focuses in one part on improving the usability of user authentication systems with graphical passwords which are believed to be easier to remember by users. His research also focuses on behavioral biometrics where mouse dynamics are used in order to determine a user's demographics such as age or gender in order to increase security against malicious users who lie about their identity. He also holds an M.S. degree from the College of William and Mary, and a B.S. degree from Geneva College.
Along the security chain of Internet communications, human users are the weakest link and are vulnerable to Internet fraud and identity theft. User authentication is a critical process of identifying a user and verifying that the user is allowed to access some restricted service and information. An effective user authentication mechanism provides a powerful guard against unauthorized access to systems and data.
In this dissertation proposal, we first propose a graphical password scheme called GridMap, in which maps are used to increase the memorability of a password for users. Moreover, GridMap employs a keyboard based input system for an increased resistance to malware and shoulder surfing attacks. A user study shows that GridMap works well in domains in which a user logs in on a regular basis, and provides a memorability benefit if the chosen map has a personal significance to the user.
To assure a user's actual identity, we propose to use mouse dynamics to determine a user's gender. In this study, mouse movement data was gathered from 96 participants and was used to identify which metrics can provide the largest differences in gender. Using the chosen metrics, we build a classifier for accurate gender identification. In our future work, we will use a similar method to determine a user's age based on mouse dynamics. A larger set of users will be used in this study in order to identify the metrics most useful for age classification, and a model will be built to attempt to determine if a user is 18 years of age or not.
Bio:
Nicolas Van Balen is a fifth year Ph.D. student at the College of William and Mary under the supervision of Dr. Wang. His research focuses in one part on improving the usability of user authentication systems with graphical passwords which are believed to be easier to remember by users. His research also focuses on behavioral biometrics where mouse dynamics are used in order to determine a user's demographics such as age or gender in order to increase security against malicious users who lie about their identity. He also holds an M.S. degree from the College of William and Mary, and a B.S. degree from Geneva College.
Contact
[[vlthompsondopp]]