Talk Info: Yiding Ran

Module Recommendation in Tertiary Education Institutions

Abstract: In this talk, Yiding will share about her work on module recommendation system. As a niche area of recommendation system, module recommendation differs from the traditional recommendation scenarios like movie and product recommendation. Other than students’ interests, factors including predicted grades also influence students’ module choices, which are often not captured by traditional recommenders. Besides assessing the performance of general recommenders on the task of module recommendation, this sharing will introduce a new model ––Grade-enhanced Multi-Layer Perceptron (GEMLP), which is designed specially for module recommendation task that balances two dimensions of students’ concerns.

Bio:

Ran Yiding is a Year 4 Business Analytics student in NUS. She started her research on module recommendation system two years ago as a UROP (Undergraduate Research Opportunity Program) student. During the one-year project, she studied how module recommendation differs from other recommendation scenarios and proposed Grade-enhanced Multi-Layer Perceptron Model for more accurate recommendation. Currently, she is working on her FYP which explores the effectiveness of KNN models in module recommendation.