Site Search
Computer Science

Photos

Xiaobu Yuan, Ph.D.Dr. Xiaobu Yuan
Dr. Xiaobu Yuan
Arunita Jaekel, Ph.D.Dr. Arunita Jaekel
Dr. Arunita Jaekel
Jessica Chen, Ph.D.Dr. Jessica Chen
Dr. Jessica Chen
Imran Ahmad, Ph.D.Dr. Imran Ahmad
Dr. Imran Ahmad
Dr. Luis RuedaDr. Luis Rueda
Dr. Luis Rueda
Windsor WaterfrontWindsor Waterfront Park
Windsor Waterfront Park
Robin Gras, Ph.D.Dr. Robin Gras
Dr. Robin Gras
Dr. Scott GoodwinDr. Scott Goodwin
Dr. Scott Goodwin
Dr. Ziad Kobti lecturingDr. Ziad Kobti
Dr. Ziad Kobti
Alioune Ngom, Ph.D.Dr. Alioune Ngom
Dr. Alioune Ngom
Dr. Robert KentDr. Robert Kent
Dr. Robert Kent
Lambton TowerLambton Tower
Lambton Tower
Christie Ezeife, Ph.D.Dr. Christie Ezeife
Dr. Christie Ezeife

Behavioural Based Biometrics Using Keystroke Dynamics for User Authentication

Add this event into your calendar using the iCAL format
  • Tue, 05/01/2018 - 11:30am - 1:30pm




Behavioural Based Biometrics Using Keystroke Dynamics for User Authentication

MSc Thesis Defense by:

Emamuzo Ogemuno

Date:  Tuesday, May 1st, 2018
Time:  11:30 am – 1:30 pm
Location: 3105, Lambton Tower

Abstract: Security of data in the recent times has become paramount, which has led to the development of many security systems. Among such system is Keystroke dynamics. Keystroke dynamics has become an active area of re-search in recent times. This is due, in part, to the increased importance of cyber-security, computer or network access control. Also known as typing dynamics, Keystroke refers to a method which identifies users/individuals based on the manner of their typing pattern or rhythm on the keyboard, which could either mean a user is verified (identified) or authenticated.  User identification is a critical factor before authentication. Now with a person already identified, the next step is to authenticate. Even if the user types in a correct password, that does not mean that the user is whom they say they are. The focus of this thesis is on the dynamic approach of keystrokes. In this work, we propose a method which improves our classification algorithm. We introduce a method that uses the Minimum Redundancy Maximum Relevance feature selection method which selects the best features based on their relevance and redundancy, we choose the best set of 10 and 20 features and performed classification tasks on this new feature sets. In addition to the feature selection, we also incorporated m-fold cross validation. This is particularly important because it gives a good representation of each class attribute in the specified fold during cross validation. We have also used the Support Vector Machine, K-Nearest Neighbour, Naïve Bayes and Grid Search for optimizing Support Vector Machine algorithm. Results not only show the efficiency of our method but also show that the proposed method can be applied to other datasets to produce optimal results.   

 

Thesis Committee:
Internal Reader:  Dr. Sherif Saad Ahmed    
External Reader:  Dr. Huapeng Wu  
Advisor:  Dr. Luis Rueda
Chair: Dr. Alioune Ngom      



csgradinfo@uwindsor.ca
(519)253-3000


See More: