A proposed new system could complement existing security safeguards by analyzing mouse movement to identify a fraudster.
While online fraud prevention has come along leaps and bounds since the early days of the internet, hackers continue to find more sophisticated ways to break into people’s online accounts.
To add another layer of security, Pranay Dave, director of AI and data science at Teradata, has published a new system that analyzes mouse movement to identify a fraudster. In his view, this could act as a passive security feature, which the user would not have to interact with.
In a blog post on Medium, Dave demonstrates how movement time, deviation, velocity, and screen distance metrics differ considerably between normal and fraud usage. In almost all instances, the unwanted user is more efficient, quicker, and doesn’t have as many outliers (sudden speedy movements).
The data comes courtesy of a whitepaper published by the Institution of Engineering and Technology, and authored by Margit Antal and Elöd Egyed-Zsigmond.
“Different ways of protection mechanism such as digital autogenerated passwords, fingerprint, or advanced techniques such as voice or facial recognition have been used [in the past],” said Dave. “Unfortunately, they are, on the one hand, intrusive and on the other hand they do not provide continuous protection.”
Dave sees mouse monitoring as a potential solution for enterprise customers, to work alongside fraud prevention solutions already in place. “The data such as distance, speed, angle combined with elementary physics knowledge can be used for creative feature engineering to develop predictive models to combat online fraud,” said Dave.