Mining Data From Interactions With a Motivational-Aware Tutoring System Using Data Visualization

Tutoring systems are a common tool for delivering educational content and recent advances in this field include the detection of and reaction to learners' motivation. A data set derived from interactions in a tutoring system and its motivationally-aware variant provided opportunities to discover patterns of behavior in connection with motivational feedback. The data collected consists of individual log files capturing the behavior of the learner during his/her interaction with the system. To mine this data, techniques were employed to discover patterns of interest when motivational scaffolding was provided by the tutoring system. A graph was constructed to visualize these patterns and to identify significant transitions derived from dyads of actions. This is a first step towards analyzing behaviors when motivational scaffolding is provided in a tutoring system. Work for the future consists of investigating the patterns' impact on learning with the motivationally-aware tutoring system.