Applied Machine Learning for Malware and Network Intrusion Detection

Wedge Networks is a leading cybersecurity solution provider in Canada. In this project, we aim to investigate the application of statistical machine learning and deep learning to cyber threat detection, aiming to detect both network intrusions and malware binaries transmitted in the network. Based on the big data collected from Wedge’s system logs and anonymized domain-specific data gathered from the clients of Wedge Networks distributed worldwide, we will investigate: 1) Distributed Denial-of-Service prevention and network intrusion detection based on both supervised and unsupervised machine learning techniques, and 2) shallow and deep neural network models for malware detection and prevention. To scale up to the big data at Wedge Networks, we will implement the developed machine learning and deep learning algorithms on distributed processing platforms such as Spark and TensorFlow. We will also integrate the learning-based threat detection module in the WedgeARP product line.

Faculty Supervisor:

Di Niu

Student:

Rui Zhu

Partner:

Wedge Networks Inc

Discipline:

Engineering - computer / electrical

Sector:

Information and communications technologies

University:

Program: