Modeling micro-interactions in self-regulated learning: A data-driven methodology

We explore whether interactive navigational behaviours can be used as a reliable and effective source to measure the progress, achievement, engagement of learning process. To do this, we propose data-driven methodology involving sequential pattern mining thematic analysis low-level interactions. applied method on an online platform which involved 193 students resulting in six that are significantly associated with learner achievement including exploration first week’s materials forum. The value these predictive models increased their explainability by 10% accounted for overall 82%. Performance evaluations indicate 91–95% accuracy identifying low-achieving students. Other relevant findings strong association between reduction over time student achievement. This suggests relationship interface learnability achievement: achievers become more efficient at using functionalities platform. These provide context progress theoretical foundations interventions against unhelpful behaviours.