An architecture for identifying and using effective learning behavior to help students manage learning

Self-regulated learners are successful because of their ability to select learning strategies, monitor their learning outcomes and adapt them accordingly. However, it is not easy to measure the outcomes of a learning strategy especially while learning. We present an architecture that allows students to gauge the effectiveness of learning behavior after the learning episode by using an interface that helps them recall what transpired during the learning episode more accurately. After an annotation process, the profit sharing algorithm is used for creating learning policies based on students’ learning behavior and their evaluations of the learning episode’s outcomes. A learning policy contains rules which describe the effectiveness of performing actions in a particular state. Learning policies are utilized for generating feedback that informs students about which actions could be changed or retained so that they can better adapt their behavior in future learning episodes. The algorithms were also tested using previously collected learning behavior data. Results showed that the approaches are capable of building a logical learning policy and utilize the policy for generating appropriate feedback.