Machine Learning Solutions for Detecting Suspicious and Abnormal User Access Roles and Entitlements

The latest cybersecurity incidents (e.g., Desjardinds, CapitalOne) have shown that the current cybersecurity solutions are not always effective in capturing and preventing attacks, especially when it comes to insider attacks that originate from within the targeted organization. In this project which will be conducted in collaboration with Xpertics, we aim to build a new solution for extracting and analyzing users’ access and entitlement data within a cloud environment, in order to detect suspicious and abnormal access activities and permissions. We will capitalize on the strength of deep learning to build strong prediction models with high accuracy. At the end of this project, the company will have an effective cybersecurity solution that could be used to prevent unauthorized access to its resources and identify the suspicious access attempts that try to manipulate the internal security measures of the company.

Faculty Supervisor:

Jamal Bentahar;Omar Abdul Wahab

Student:

Partner:

Xpertics Solutions Inc

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

Concordia University

Program: