Academic and STEM career pathway recommendation using video gameplay data

This project involves using video gameplay data to generate personalised course recommendations and career pathways in STEM. The project will utilise open-source large language models (LLMs) on the data collected from video gameplay to recommend courses and career pathways. The project will use Transformer-based language models and recommender systems grounded in a human behavioural research. The project is in collaboration with Arludo.

Arludo is an education and research company whose mission is to excite students about learning STEM topics, with the goal of increasing the diversity of individuals within STEM fields. Our approach involves creating science-based video games where students discover concepts through play. As they play, they also collect science data that are visualised in front of the classroom in real time. Each Arludo game also collects behavioural data. Our primary goal is to use the data collected from videogames along with data from student answers to better understand student personality and suggest courses and career pathways that match student interests. A secondary goal is to identify student personalities, and recommend games that will broaden student horizons to help students understand that they belong, and can succeed, within STEM fields. Throughout this project, you will be working with leading natural language processing (NLP) experts as well as experts in human behaviour and our industry partner (Arludo) to solve real-world problems.