Analysing the Learning Paths of Computer Science Students via Trajectory Analytics

Universities collect an enormous amount of student data that remains underutilized in helping decisions or policymaking. Strategies to deal with problems such as drop-out students and low achievers are imperative to the success of universities. Our work is motivated by the fact that student journey in a learning environment, such as a university, resembles a trajectory. The journey of the student is analogous to that of a vehicle trajectory, in which a student is similar to a vehicle which passes through different units/subjects instead of different geographical locations to reach a certain graduate outcome (destination). The trajectory concept provides a model to capture and represent a large amount of data collected by universities in a meaningful way. In this paper, we develop a trajectory model to capture the journey of students who are enrolled in Computer Science courses. We focus on several modelling aspects, including trajectory representation and trajectory clustering that can be applied to gain valuable insights on students' journeys through university, both at individual and collective levels.