Some researchers have developed relevant and diverse proposals for improving the content quality of the learner model in Intelligent Tutoring Systems, mainly reducing its uncertainty. Following
this aim, this paper proposes an open learner modeling approach using
Bayesian networks, focusing on negotiation mechanism to solve detected
cognitive conflicts that can emerge when the learner inspects information
of his model inferred by the system. Therefore, we addressed some issues
concerning the provision of inspectable model and negotiated updating
of this model. Its contribution lies in the fact that the learners attempt
to change the learner model is met with a challenge, leading to a decision
if the learner claims to know more (or less) than the model represents.
Testing tools and AI - ideas what to try with some tool examples
An Open and Inspectable Learner Modeling with a Negotiation Mechanism to Solve Cognitive Conflicts in an Intelligent Tutoring System
1. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
An Open and Inspectable Learner Modeling with a
Negotiation Mechanism to Solve Cognitive Conflicts
in an Intelligent Tutoring System
Evandro Costa, Priscylla Silva,
Jonathas Magalh˜es and Marlos Silva
a
TIPS Group
Computing Institute
Federal University of Alagoas, Brazil
Federal University of Campina Grande, Brazil
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 1
2. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Research Context
Learner modeling tasks in ITS;
High level of uncertainty;
Probabilistic Learner Modeling in ITS;
Opening and Viewing Learner Model;
Presence of Cognitive Conflicts;
Mechanisms for dealing with conflicts;
Negotiating the open learner model.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 2
3. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Research Questions
Q1 : What approach should we adopt to deal with uncertainty
found in a learner model for ITS?
Q2 : What is an appropriate way to define and viewing OLM?
Q3 : How can we detect cognitive conflicts between the student
and the system concerning problem solving activities?
Q4 : How can we effectively address these conflicts?
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 3
4. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
How those Questions have been addressed?
With respect to Q1 – Representation and Maintenance:
Conati et al. [2];
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 4
5. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
How those Questions have been addressed?
With respect to Q2 – OLM and Visualization:
Zapata and Greer [5];
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 5
6. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
How those Questions have been addressed?
With respect to Q3 and Q4 – Conflicts detection and Negotiation:
Bull et al. [1];
Dimitrova [3];
Thomson and Mitrovic [4].
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 6
7. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Our General Approach
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 7
8. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
The Open Learner Model
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 8
9. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
The Open Learner Model
1 4
The system put a problem to the learner: 3 + 3;
He declares his belief:
Very unsure = 0.05;
Unsure = 0.25;
Almost sure = 0.5;
Sure = 0.75;
Very sure = 0.95.
Then, he submits a solution and the system evaluate it and
returns a grade [0,1].
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 9
10. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
The Open Learner Model
(a) The Ms . (b) The Mt .
Figure: Task-specific part of the Learner Model.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 10
11. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
The Open Learner Model
(a) The Ms . (b) The Mt .
Figure: Domain-general part of the Learner Model.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 11
12. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
The Open Learner Model
Figure: The Visualization of the Learner Model.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 12
13. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Negotiation Process
The negotiation mechanism depends on the learner’s credibility:
Figure: The DBN of the learner’s credibility.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 13
14. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Negotiation Process
When the learner wants to change the tutor’s belief
Credibility L’s belief < T’s belief L’s Belief > T’s belief
Low Persuasion Persuasion
Medium Persuasion Cooperation
High Persuasion Cooperation
When the tutor wants to change the student’s belief
Credibility L’s belief < T’s belief L’s belief > T’s belief
Low Support Contestation
Medium Support Contestation
High Support Contestation
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 14
15. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Prove Process
During the negotiation:
The system can request that the learner proves his knowledge, or;
The learner can request the opportunity of prove.
The proof process consists of:
Two problems and the learner has two chances to solve each
problem;
Then, his model is updated.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 15
16. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Demonstration Scenario
Figure: Example of Negotiation Dialogue Started by the Learner.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 16
17. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
We are we going next?
Improve the visualization, allowing the visualization of the two
parts of the model;
Put other evidences in the learner model: social characteristics,
CV-curriculum of the student, collaborative information;
Perform an experiment in a basic math classroom.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 17
18. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
References
Susan Bull, Paul Brna, and Helen Pain.
Extending the scope of the student model.
User Modeling and User-Adapted Interaction, 5(1):45–65, 1995.
Cristina Conati, Abigail Gertner, and Kurt Vanlehn.
Using bayesian networks to manage uncertainty in student
modeling.
User Modeling and User-Adapted Interaction, 12(4):371–417,
2002.
Vania Dimitrova.
Style-olm: Interactive open learner modelling.
International Journal of Artificial Intelligence in Education,
13(1):35–78, January 2003.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 18
19. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Referˆncias
e
David Thomson and Antonija Mitrovic.
Preliminary evaluation of a negotiable student model in a
constraint-based its.
Research and Practice in Technology Enhanced Learning,
5(1):19–33, 2010.
Juan-Diego Zapata-Rivera and Jim E. Greer.
Interacting with inspectable bayesian student models.
International Journal of Artificial Intelligence in Education,
14(2):127–163, 2004.
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 19
20. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
Thanks!!
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 20
21. Introduction
Related Work
Our Proposal
Demonstration Scenario
Final Considerations
References
For more information: http://tip.ic.ufal.br/site/
E. Costa, P. Silva, J. Magalh˜es and M. Silva
a PALE UMAP 2012 21