Towards Integrating Human and Automated Tutoring Systems
暂无分享,去创建一个
We envision next generation learners having access to both automated and human sources of instruction in a variety of learning contexts. In such contexts, it will be most effective if students can be assisted to appropriately navigate between these sources of instruction. For example, human tutors, when helping a struggling student, might benefit from having access to the learning profile an automated tutor possesses on the student, including what the student already knows, detected misconceptions, inferred affective state and details about the student's work with the automated system before requesting human help. Similarly, an automated tutoring system would benefit from knowledge of interactions during human tutoring session. To facilitate student transitions between these types of systems, we need to understand the factors that best aid students in transitioning between such systems. This poster reports preliminary analyses, suggesting that students who are struggling with the course are more likely to take advantage of the optional human tutoring support and that such use is associated with increased course completion rates, regardless of the student’s level of preparation.
[1] Steven Ritter,et al. Predicting Standardized Test Scores from Cognitive Tutor Interactions , 2013, EDM.
[2] Arthur C. Graesser,et al. Recent Advances in Conversational Intelligent Tutoring Systems , 2013, AI Mag..
[3] Albert T. Corbett,et al. Cognitive Tutor: Applied research in mathematics education , 2007, Psychonomic bulletin & review.