EduMark AI

The EduMark project was funded by the Drapers' Fund for Innovation in Learning and Teaching 2024/25. It aims to leverage AI to provide faster, more consistent grades, personalised student feedback and time-saving for educators.

Project approach

1. Conducted comparative analysis of multiple AI systems (ChatGPT, Google Gemini, Graide, Timely Grader)
2. Ran student surveys to capture qualitative feedback (200+ students engaged)
3. Collected paired educator vs. AI grading samples for benchmarking
4. Developed and launched a beta EduMark AI web app with an intuitive interface

Project deliverables

1. EduMark AI beta web app (edumark-ai.com)
2. Statistical analysis showing 50–60% reduction in grading time compared to manual marking
3. Evidence of higher consistency in grade distributions and rubric alignment
4. Student survey results: overwhelmingly positive, praising specific, actionable feedback
5. Educator workflows including “AI + educator oversight” hybrid model to maintain academic integrity

The project has been disseminated through blogs (JISC, Advance HE, Queen Mary Centre for Excellence in Artificial Intelligence in Education) and poster presentations  (Queen Mary Festival of Education and Royal Society of Chemistry Poster Presentation).

It was also selected by OpenAI Academy for their international "Professors Teaching with AI" series.

Project team

Deepshikha Deepshikha (School of Engineering and Materials Science; Conrad Bessant (School of Biological and Behavioural Sciences); Darryn Mitussis (School of Business and Management); Li Wang (School of Engineering and Materials Science);  Xinru Deng (School of Engineering and Materials Science); Giuseppe Viola (School of Engineering and Materials Science) and Mouna Chetehouna (School of Engineering and Materials Science).