Cohesion-based Analysis of CSCL Conversations: Holistic and Individual Perspectives
Résumé
Although Computer Supported Collaborative Learning (CSCL) technologies have gained an increasing role in educational environments, there are few automatic systems that address involvement, knowledge-building and collaboration in order to support tutors in the time consuming process of analyzing conversations. We propose a cohesion-based analysis model integrating multiple natural language techniques, an intervention scoring mechanism and a comprehensive collaboration assessment method, derived from social knowledge-building, reflected at utterance level through cohesion. Furthermore, by combining a holistic perspective of the entire conversation with a more fine grained view focused on each participant, we obtain a thorough evaluation of chat conversations with focus on topics modeling, participant interaction and collaboration. In order to sustain our model, we have performed a preliminary validation study that proves that our analysis is consistent with tutor evaluations.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...