Optimization of Personalized Learning Pathways Based on Competencies and Outcome

Abstract

Personalized learning that is tailored to individual needs, preference, and interests may improve student learning experience and outcome. With the aid of computing technology, it is becoming possible to deliver personalized learning to a large and diverse student population. One of the key problems involved is the determination of the pathway which each learner follows to complete a learning program. Existing methods generally rely on a priori knowledge of subject content prerequisite relationship or constraint to determine the sequencing of instructional materials without any consideration of student learning outcome. In this paper, we formulate the selection of learning pathways as an optimization problem based on competencies and student learning outcome. We show that the resulting pathway selection problem can be modeled as a Markov Decision Process (MDP). Decision rules can thus be defined and applied to select personalized learning pathways to optimize student learning outcome according to desired performance criteria.