Transcript upload:Gerardo_Presentation

VMT Workshop
june 9-11, Philadelphia
Gerardo Ayala
Centro de Investigación en Tecnologías de la Información y
Automatización, CENTIA
Universidad de las Américas, Puebla
MEXICO
GRACILE






Learner modelling and software agents for CSCL
environments.
Groups of foreign students learning Japanese.
The learner model in a CSCL environments supports
awareness and enhance the relevant collaboration between
learners.
The learner model is considered a set of beliefs held by a
mediator agent about the learner's goals, commitments and
capabilities.
The mediator agent maintains a zone of proximal
development for its learner, based on the structural and the
social aspects of knowledge progress.
A domain agent assists the learners in their constructions.
GRACILE

Agents maintaining zones of proximal development
(structural and social knowledge frontiers)
GRACILE

Open learner models
HTML Document
Gracile.gif
GRACILE




Ayala, Gerardo & Yano, Yoneo. (1998) A Collaborative Learning
Environment Based on Intelligent Agents, Expert Systems with
Applications, Pergamon Press, pp: 129-137
Ayala, Gerardo & Yano, Yoneo. (1997) Evaluating the Performance
of Agents that Support the Effective Collaboration of Learners in
a CSCL Environment, IEICE Transactions on Information and
Systems, Special Issue on Educational Systems Using Multimedia and
Communication Technology,
February 97, Vol.E80-D, No.2, pp: 35HTML Document
Gracile.gif
46.
Ayala, Gerardo. & Yano, Yoneo. (1996). Learner Models for
Supporting Awareness and Collaboration in a CSCL Environment,
Intelligent Tutoring Systems, Claude Frasson, Gilles Gauthier and Alan
Lesgold (Eds.), Lecture Notes in Computer Science 1086, Springer
Verlag 1996, pp: 158-167
Ayala, Gerardo. & Yano, Yoneo. (1995). GRACILE: A Framework
for Collaborative Intelligent Learning Environments, Journal of the
Japanese Society of Artificial Intelligence, Vol.10. No. 6. pp: 156-170.
CASSIEL
• Agent supporting lifelong learning.
• Communities of professionals, university students and
researchers.
• Enculturation.
• Maintains individual and group competitive advantage.
• Personalized learning plan and group configuration, using a
learner model server based on the interests of the learner.
• Implementing a theory of social construction of knowledge,
from the area of knowledge management.
CASSIEL
CASSIEL



Ayala, Gerardo. (2001) Intelligent Agents Supporting the Social
Construction of Knowledge in a Learning Environment, Human
Computer Interaction: Issues and Challenges, (Qiyang Chen, Editor.)
Idea Group Publishing, Chapter 4, pp. 44-63
Ayala, Gerardo. (2002) Intelligent Agents for a Lifelong Learning
Environment in Information Technologies, International Journal of
Continuing Engineering Education and Life-Long Learning, Vol. 12,
Nos. 1-4, pp. 189-200.
Ayala, Gerardo. (2003) Towards Lifelong Learning Environments:
agent supporting the collaborative construction of knowledge in
virtual communities, Designing for change in networked learning
environments, proccedings of the International Conference on
Computer Support for Collaborative Learning CSCL 2003, Bergen,
Norway, pp. 141-148.
Current research interests
Learner modelling for CSCL environments
 Zones of proximal development.
 Group configuration for effective collaboration.
 Motivation based on personalization and
awareness.
 Answer set programming for belief revision in
learner modelling.
Agents supporting lifelong learning
 Large communities of users.
 Social construction of knowledge.
 Second language learning.
 Social practices.
How to attract and motivate the
participants of a CSCL environment?
• Try to maintain always some learning possibilities for all.
• Feeling of being useful and productive in the community.
• Awareness of mutual benefit by collaborating.
Concept awareness

Reflection of the learner about her/his current level with
respect to the others
Feeling of learning something useful, relevant and new,
by collaborating.
How to make the learning community
grow and mature?
• Communities grow and mature if there is a
genuine common interest.
• Reputation: Can students express to others what
did they learn (benefits) by participating in the
environment?
•Added Value to the environment (something plus
the chat with the WYSIWIS graphic interface)
• Keep it Fun!
How to group participants for an
effective collaboration?
Group configuration, based on learner modelling.
 Capacities
 Commitments
 Interests
Keep zones of proximal development for all.
 Task asignment based on the interests and the zone
of proximal development of the participant in a
given group.
This requires a structure of the domain
How to design problems that will foster
group and individual understanding?
 Are
the problems and the environment more
oriented to collaborative problem solving
rather than collaborative learning?
public class CSCL extends CSCW
{
….
}
How to design software that supports
effective collaboration?
Social awareness
 Appropriate group configuration for learning
Workspace awareness
 Include a personal workspace in the interface
 Space for agreements in the interface
Concept Awareness
 Feedback (did I solve the problem?)
Task awareness
 Clear relations between a given situation in the problem
and the concepts involved (learning outcome).