An Evidence-Based Learner Model for Supporting Activities in Robotics

An Evidence-Based Learner Model for Supporting Activities in Robotics

연구 분야: Artificial Intelligence

학회: L@S '20: Proceedings of the Seventh ACM Conference on Learning @ Scale

Teaching robotics is an attractive way of motivating students to learn computer science. However, it is also a challenging topic for students of all ages and only one teacher in a classroom is too little to support approximately 30 students at the same time. Therefore, intelligent tutoring systems might be a meaningful way to support students and teachers. In this paper we describe an approach to support computer science lessons in secondary schools by using a learner model. We are explaining how the three phases of our learner model (data collection - profile construction - profile application) can be implemented for teaching robotics by using different types of implicit and explicit data to generate feedback for the teacher concerning competencies and knowledge of the students on the one hand and by supporting collaboration and group formation amongst the students on the other hand. The model is derived from literature and supported by data from different studies.