HQ Trivia Bot

My final year computer science project dissertation was "Quickly Answering Multiple-Choice Questions using Deep Learning", which involved creating an automatic answering bot the (then) popular game HQ Trivia. The area of study involved performing open-domain question answering on games of HQ Trivia, using deep learning and Google AI Language's BERT to generate answer predictions in real time. This was a bleeding-edge deep learning model, amongst the first to use a transformer model at its core. This was my first exploration into machine learning and found it extremely interesting; I was able to achieve a result similar to the state of the art in open-domain, multiple-choice question answering. You can see the full paper here.

I attended Durham University from 2016-2019 and graduated with First Class Honours in BSc Computer Science.