Identity Fraud Detection through Anomalous Login Patterns Mining

Identity fraud is spreading fast and causing more and more damages both financially and sociologically. Identity fraud occurs when a criminal impersonates another individual by taking on that person’s identity or by creating a fake identity for whatever reason. The project will investigate and develop a new model for identity fraud detection based on login sequence, history and context. Analysis of such information using data mining will allow tracking individual login behaviors and identifying anomalies and inconsistencies in login occurrences that potentially point to fraudulent activities. The outcome of the proposed research will be implemented as an add-on to the existing flagship authentication platform of the partner organization.

Faculty Supervisor:

Mihai Sima

Student:

Amany Abdelhalim

Partner:

Plurilock Security Solutions Inc.

Discipline:

Engineering - computer / electrical

Sector:

Information and communications technologies

University:

University of Victoria