uOttawa student develops machine learning tool to combat credit fraud

With the number of credit card transactions rapidly increasingly by the day, one University of Ottawa student is doing what she can to tackle credit card fraud with advanced machine learning techniques.

Bahar Emami Afshar, a computer science master’s student, has created an advanced machine learning program, designed specifically for the financial sector, that will be able to detect anomalies in large datasets.

“Credit card fraud has become a critical issue in today’s world with the rapid growth of e-commerce and online transactions,” Afshar said in a media release. “What makes addressing it even more urgent is that this type of fraud is often tied to broader illicit activities, such as identity theft and money laundering.” 

In 2022, Canada saw 20.5 billion payment transactions, according to Payments Canada. The scale of this dataset would make it near impossible to identify manually.

Afshar’s ITERADE (ITERative Anomaly Detection Ensemble) tool, prioritizes labelling efforts to maximize fraud detection.

Testing of the ITERADE tool has shown some preliminary success. Afshar and her collaborators have seen fraud detection rates three to 15 times better as a result of this tool than those for previous tactics.

That success is important as credit card fraud remains a grave issue facing the business sector. A report from Payments Canada published in September 2024 found that in the previous six months, one in five Canadian businesses had experienced payment fraud. Twenty per cent of these cases were credit card fraud, the third leading cause.

“Addressing fraud risks is a central focus for the payment ecosystem. It requires a multifaceted approach that leverages technology, system innovations, evolving regulations and education through continued industry collaboration,” Donna Kinoshita, Chief Payments Officer at Payments Canada, said in a press release.

That is exactly what Afshar’s tool strives to do, and its success could prove beneficial for other sectors as well.

“In the research community, ITERADE introduces a novel unsupervised anomaly detection approach, which shifts how we address highly imbalanced datasets. It provides a flexible, budget-agnostic framework that outperforms existing methods in fraud detection and sets a new standard for handling other imbalanced classification problems in an unsupervised manner,” Afshar said.

As her work continues, Afshar and her team are looking to integrate ITERADE into an active learning framework which could solve challenges beyond the finance sector, in industries like health care, cybersecurity and infrastructure monitoring.

“I’m excited to pursue a career that combines AI and engineering to tackle complex real-world problems,” Afshar said. “I aim to contribute to cutting-edge AI applications that improve decision-making, security and overall quality of life across various industries.”