For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Evaluating the Performance of Agents that Support the Effective Collaboration of Learners in a CSCL Environment
Gerardo AYALA San Martin Yoneo YANO
IEICE TRANSACTIONS on Information and Systems
Publication Date: 1997/02/25
Print ISSN: 0916-8532
Type of Manuscript: Special Section PAPER (Special Issue on Educational System using Multimedia and Communication Technology)
Category: Collaboration and Agent system for learning support
CSCL environments, effective collaboration , knowledge negotiation, intelligent agents,
Full Text: PDF>>
Effective collaboration in ComputerSupported Collaborative Learning (CSCL) environments is nowadays an important research topic. It deals with two main problems: the configuration of an appropriate learning group and the intelligent task distribution in the practice of domain knowledge. In order to have effective collaboration in a CSCL environment, we have proposed a set of software agents that assist the learners to select their learning tasks, according to their capabilities and the possibilities of collaboration between them. In this paper the cooperation among software agents is presented as the key point for effective collaboration in CSCL environments. In this kind of environments the learner must have enough collaboration and learning possibilities, being motivated with the experience of social knowledge construction. We have been working on the problem of effective collaboration in CSCL environments, based on the cooperation between software agents developed for GRACILE, our Japanese Grammar CSCL environment. Before, we have proposed intelligent agents that assist the learners. Our next step has been the design of the cooperation between agents in order to create possibilities of effective collaboration in a virtual community of practice. In order to evaluate the performance of our agents we made several simulations. The results obtained from these simulations of diverse types of learning groups provided us with guidelines for the configuration of groups in CSCL environments, where effective collaboration is possible.