PROJECT TITLE :

A Machine Learning Approach for Tracking and Predicting Student Performance in Degree Programs - 2018

ABSTRACT:

Accurately predicting students' future performance based on their ongoing academic records is crucial for effectively carrying out necessary pedagogical interventions to make sure students' on-time and satisfactory graduation. Although there's a rich literature on predicting student performance when solving issues or learning for courses using data-driven approaches, predicting student performance in finishing degrees (e.g., college programs) is a lot of less studied and faces new challenges: (one) Students differ tremendously in terms of backgrounds and selected courses; (two) courses don't seem to be equally informative for creating accurate predictions; and (three) students' evolving progress needs to be incorporated into the prediction. In this Project, we tend to develop a novel machine learning technique for predicting student performance in degree programs that is in a position to address these key challenges. The proposed methodology has two major options. First, a bilayered structure comprising multiple base predictors and a cascade of ensemble predictors is developed for creating predictions based mostly on students' evolving performance states. Second, a data-driven approach based mostly on latent issue models and probabilistic matrix factorization is proposed to discover course relevance, that is very important for constructing efficient base predictors. Through intensive simulations on an undergraduate student dataset collected over three years at University of California, Los Angeles, we have a tendency to show that the proposed technique achieves superior performance to benchmark approaches.