Improving pedagogical recommendations by classifying students according to their interactional behavior in a gamified learning environment

Abstract

In this work we present a way to classify students from a gamified online learning environment according to their interactions, and to use the results to create and recommend "missions" - a gamification element that contains challenging tasks to keep students engaged - that focus on: (1) the students' most common interactions, (2) the students' least common interactions (to balance their online behavior), and (3) more than one type of interaction at the same time. The classification approach was systematically applied, following the Pedagogical Recommendation Process. It's main objective was to assist teachers creating personalized missions. We applied a questionnaire to compare the new approach (classification according to patterns in interactions) to the existing one (a report showing some actions students took in the environment). In the results, the classification approach provided a better way to evaluate the students, regarding the three missions' focus.