Studying how technology changes what people can do.

FakeForward: Using Deepfake Technology for Feedforward Learning

Christopher Clarke, Jingnan Xu, Ye Zhu, Karan Dharamshi, Harry McGill, Stephen Black and Christof Lutteroth

The question

Video self-modelling can help a person learn by showing them performing a skill at a level they have not yet reached. Traditional approaches require extensive recording and manual editing. FakeForward asks whether face-swapping can create a useful self-modelling video from footage of someone who already performs the skill well.

What the research covered

The team studied physical exercises and public speaking. The work looked at performance as well as confidence, anxiety and perceived competence, placing the technical method inside a wider question about learning and motivation.

Why it matters

The paper shows how a technical technique can be treated as a learning tool rather than an end in itself. It also raises important questions about consent, authenticity and the responsible use of altered video.

Chrome Extension for Detecting Phishing Websites

The problem

Blacklist-based detection can miss a new phishing address simply because it has not yet been recorded. The project explored a different approach: classifying the structure of the URL itself.

The approach

We compared machine-learning methods, selected a Random Forest model and built the result into a Chrome extension. The extension evaluates features from a web address and classifies the page as likely legitimate or phishing.

The paper was published in the International Research Journal of Engineering and Technology, Volume 7, Issue 3.