A case-based peer-to-peer framework for managing student models in Intelligent Tutoring Systems

Intelligent tutoring systems (ITSs) aim at providing personalized and adaptive tutoring to students by the incorporation of a student modeling component. In the near future, a very interesting scenario will appear when multiple tutoring systems exchange information in order to learn from its own experiences and improve their student modeling components. In order to get closer to such scenario, in this paper we present a case-based peer-to-peer multi-agent system for collaborative management of student models in ITSs. The goal of the system is twofold: first, to initialize the student model when a new student logs on the tutor system and second, to update the student model depending on the studentpsilas interaction with the system and exchanging this information with its peers. The quality of the system is evaluated in terms of its ability for searching similar cases (accuracy) tested under three different strategies. Our results show that increasing the system complexity (number of nodes and/or number of students) and using a committee strategy, the performance of the global system is improved by reducing network traffic, and preserving the quality of the solutions for the new students (cases).