Software

Alvis

Belvedere

COLER

disCourse

Kukakuka

Pink

 Outreach
 Opportunities
 Sponsors

email us

LILT, Dept. of ICS • University of Hawai`i at Manoa • 1680 East-West Road, POST 309 • Honolulu, HI 96822 • 1-808-956-3890

COllaborative Learning environment for Entity-Relationship modeling (COLER)

María de los Angeles Constantino-González
Dept. of Computer Sciences, ITESM Campus Laguna
Torreón, México
aconstan@campus.lag.itesm.mx

(Work done at ITESM. Thesis advised by Dan Suthers)

The COLER project explores ways in which an intelligent system can assist with facilitation of collaboration in a synchronous distance learning context. Support for practice of collaborative skills is essential in distance learning because remote learners may have few opportunities to collaborate with others. Facilitators may find it difficult to monitor students' collaboration and guide participation, because many teams have to be monitored at a time. COLER provides a web-based environment in which students can construct individual solutions to entity-relationship modeling problems and then collaborate with others in merging their individual solutions in a group solution. COLER includes a software agent designed to coach collaboration.

Prior work on supporting collaboration has relied largely on comparing student discourse to models of collaborative discourse. Comparison of student work to expert solutions is prevalent in individual coaching paradigms. Although these approaches are valuable, our approach evaluates the potential contribution of tracking student participation and comparing students' individual and group solutions. Following socio-cognitive conflict theory and the theory of cognitive dissonance, our theoretical motivation is that conflicts between individual and group solutions constitute learning opportunities, provided that students recognize and address these conflicts. The coach encourages such negotiation when differences are detected, and also encourages participation in other ways.

Evaluation of the system was based primarily on expert judgement, and secondary on students' reactions to coach advice. Results show that the quality of the advice was good; however, other knowledge should be included to improve coverage of advice to a broader range of situations and advice types. Evaluation of the contribution of each knowledge source indicated that the system was able to produce reasonable advice out of a few basic knowledge sources: solution differences, participation in solution construction, feedback given to other students, and chat activity.

This work makes two major contributions. First, the work defined and evaluated a new approach to coaching collaboration based primarily on tracking students' participation and recognizing differences between students' individual and group solutions. This approach can be applied to learning tasks involving group problem solving in which structured representations of problem solutions exist. The second contribution was the identification of the knowledge required by a collaboration coach, and the architecture in which this knowledge is applied. Since most of the knowledge defined for coaching collaboration is independent of the domain, this design can facilitate the creation of collaborative coaches in other domains.

Related Publications

Constantino-Gonzales, M. A., Suthers, D., Icaza, J. (2001).

Designing and Evaluating a Collaboration Coach: Knowledge and Reasoning. 10th International Conference on Artificial Intelligence in Education, May 19-23, San Antonio Texas.
PDF preprint

Constantino-Gonzales, M. A., & Suthers, D. (2001).

Coaching Collaboration by Comparing Solutions and Tracking Participation. In P. Dillenbourg, A. Eurelings, K. Hakkarainen (Eds.) European Perspectives on Computer-Supported Collaborative Learning, Proceedings of the First European Conference on Computer-Supported Collaborative Learning, Universiteit Maastricht, Maastrict, the Netherlands, March 22-24 2001, pp. 173-180.
PDF preprint

Constantino, A. & Suthers, D. (2000).

A coached collaborative learning environment for entity-relationship modeling. In Intelligent Tutoring Systems, Proceedings of the 5th International Conference (ITS 2000), G. Gauthier, C. Frasson, & K. VanLehn (Eds.), pp. 325-333. Berlin: Springer-Verlag.
PDF preprint