Monitoring and analysing students' systematic behaviour - The SCY pedagogical agent framework

Abstract

In this paper we present an agent-based architecture to monitor and analyse students' behaviour while working with a simulation tool. This architecture is based on the blackboard principle and implemented using a TupleSpaces approach. The agents are divided into several groups according to their tasks to maximise autonomy and reusability. This separation was done with the idea of perceive-react-cycles in mind. An implemented example scenario that demonstrates how agents interact in generating meaningful context-aware student feedback is described. The visualisation of the agents' analysis results facilitates a deeper understanding of the learners' activities. © 2010 Springer-Verlag Berlin Heidelberg.