Agent-based collaboration systems: A case study

Experimental studies concerning distance learning show that forms of interaction like those promoted by collaborative learning lead to different and desirable cognitive activities. However, existing distance learning platforms provide collaboration tools (like forum, chat, Wiki, etc.) that present some limits that makes them mostly ineffective from the standpoint of collaborative learning. As a consequence, a new infrastructure, able to bridge the gap between the interaction forms of collaborative learning and existing learning collaboration tools, is needed. In this context, Multi-Agent System (MAS) seems to be a suitable paradigm to engineer collaborative distance learning systems for it promotes effective collaboration, overcoming the aforementioned limits. In this paper we discuss how technologies like intelligent agents and meta-models like the Agents and Artefacts (A&A) could be exploited to build a framework for collaborative distance learning. In particular, we show an architecture based on the A&A meta-model promoting the integration between heterogeneous collaboration tools and services re-casted as artefacts, both to coordinate human interactions and to allow knowledge construction and reuse according to the educational models adopted. Moreover, we show how artefact properties make it possible for both human and intelligent agents to monitor collaborative activities of human beings and give feedbacks to students and teachers based on social interaction and the specific learning status. This is an important aspect of distance learning since learners, in absence of continuous feedback, feel alone and abandoned, so they are less stimulated to learn.