Visual Module Exploration: A Live-User Evaluation

Modern universities present students with a wide array of course and module options, making it difficult for students to make informed decisions about the modules they take and how their choices can help them achieve their educational goals. This is exacerbated when students are uncertain about their goals or when limited information about module options is available, as is all too often the case, leaving many students to follow the choices of their peers. The main contribution of this work is to describe a module recommendation and advisory system to help undergraduate students better understand the options available to them and the implications of their decisions. We describe a system that uses text mining techniques on raw module descriptions to generate rich, interconnected module representations. We demonstrate how these representations can be used as the basis for a visual recommender system and describe the results of a recent live-user evaluation to demonstrate the practical benefits of such a system on different groups of undergraduate students. This paper is an extended version that was previously presented at AI-2021 Forty-first SGAI International Conference on Artificial Intelligence [ 1 ].