Matheesha Fernando (pictured above), a PhD candidate in Computer Science and Cybersecurity, is investigating phishing attacks.
“Phishing is a type of scam where a message or email is disguised as coming from a trusted sender, tricking the recipient into clicking a URL of a counterfeit website collecting sensitive information or enabling the deployment of malicious software,” she explains.
Fernando’s and her supervisors Associate Professor Abdun Mahmood and Dr Jabed Chowdhury apply Machine Learning, a form of Artificial Intelligence, to enhance phishing detection and categorisation for better awareness.
“I hope that my work helps to bridge the knowledge gap in contemporary phishing attacks and human awareness, so that internetusers can be better protected from them.”
“Cybersecurity is essential to modern living, and I am fulfilling a dream to contribute to research in this area.”
PhD candidate, Josh Millward, is investigating how machine learning can benefit cancer treatment.
“After graduating from La Trobe in 2018, I began work as a machine learning research engineer,” says Millward. “But I always knew I wanted to go back to do a PhD.”
Choosing to study at La Trobe, says Millward, was a natural choice.
“I really enjoyed my undergraduate experience at La Trobe and was excited to return. I am also working with two excellent supervisors who I know will help guide me through my PhD to completion.”
Millward’s research explores how machine learning can be used to improve digital pathology in cancer treatment.
“Digital pathology involves creating images from tissue, which allow pathologists to perform detailed analysis on a computer. In oncology, for example, this might involve counting the number of cells in a sample that are associated with cancer relapse. But, it is very time consuming to count them manually,” explains Millward.
“My research is investigating if deep learning algorithms can be used to automate this analysis. It is hoped that our findings will give pathologists and oncologists the tools to expedite patient diagnosis.”
PhD candidate, Matt Felicetti, is investigating how to enhance data modelling in industrial settings through machine learning.
“I chose to undertake a PhD because being able to teach, research, be involved in interesting projects and work with like-minded people appealed to me greatly.”
Felicetti’s research seeks to address some of the challenges faced by industry when it comes to data modelling.
“Data modelling using machine learning is difficult in industrial settings because it must be as fast as possible to match the demands of industry, and there are limited hardware resources to run it,” explains Felicetti.
“I have been investigating algorithms and implementations that produce data models that are simpler, faster and have a lower computation cost – I hope to show that this can be translated to industry.”