Did the Team Actually Collaborate? Mining Multi-Channel Signals in Group Software Projects

Project description

Group software development projects are a cornerstone of computing education, yet fairly assessing individual contribution remains a persistent challenge for educators. This project aims to explore whether collaboration signals drawn from multiple sources, such as version control histories, chat logs, and issue trackers, can reliably distinguish genuine joint work from non-contribution. The student will conduct a literature review to identify and evaluate existing contribution metrics and their limitations, before designing and implementing a cross-channel feature pipeline and evaluating it against tutor judgements of real student teams. This project will explore the relationship between multi-channel collaboration signals and single-source git-only baselines, examining how well each predicts educator-assessed contribution in group software projects.

Note: You need to register interest in projects from different supervisors (not a number of projects with the one supervisor).
You must also contact each supervisor directly to discuss both the project details and your suitability to undertake the project.