1 / 51

Visualizing Collaboration

Visualizing Collaboration. Christopher Teplovs Computational Social Science Lab (CSSL) Copenhagen Business School. Visualizing (Opportunities for) Collaboration. Christopher Teplovs Computational Social Science Lab (CSSL) Copenhagen Business School. Outline. background stakeholders

haruki
Télécharger la présentation

Visualizing Collaboration

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. VisualizingCollaboration Christopher Teplovs Computational Social Science Lab (CSSL) Copenhagen Business School

  2. Visualizing(Opportunities for)Collaboration Christopher Teplovs Computational Social Science Lab (CSSL) Copenhagen Business School

  3. Outline • background • stakeholders • students • teachers • researchers & observers • Knowledge Forum • Knowledge Space Visualizer • KISSME

  4. Background

  5. Online Discourse • product (writing, drawing, etc.) • process (who has done what and when) • focus on text-based analysis

  6. Knowledge Forum

  7. Knowledge Forum

  8. Knowledge Forum

  9. Knowledge Forum

  10. Knowledge Forum • based on principles of Knowledge Building • assessment via embedded tools • information visualization

  11. Embedded and Transformative Assessment • Standards and benchmarks are objects of discourse in Knowledge Forum, to be annotated, built on, and risen above.

  12. Participation & CollaborationTools • These tools provide information about the number and nature of artifacts created by • participants at the individual and group level, as well as relationships between individuals.

  13. Social Network AnalysisTools

  14. Writing Analysis Tools • These tools parse and quantify the contributions of participants in terms of vocabulary growth and basic writing measures (e.g. total and unique words, mean sentence length).

  15. Writing Analysis Tools(small multiples)

  16. Semantic Analysis Tools • These tools deal with the meaning of the discourse by analyzing the semantic fields of contributions.

  17. Semantic Analysis Tools

  18. Visualization Tools • These tools allow participants and observers to make sense of the emergent knowledge building discourse.

  19. The Knowledge Space Visualizer (KSV) • designed to allow researchers to use computer-assisted 2D visualizations of learner-generated contributions to online discourse • can visualize explicit links between documents • replies, references, annotations

  20. KSV: Visualizing Implicit Links • previous example showed explicit links between documents • can also show implicit links based on similarity of documents • LSA gives us vector representations of documents • cosine between vectors is a good indicator of the similarity between documents

  21. KSV: Computer-assisted layout • axes can be operationally defined to provide a variety of layout options • chronology and authorship can be used to see when participants contribute

  22. Of course we can overlay additional information, like the relationships between documents

  23. Of course we can overlay additional information, like the relationships between documents

  24. The KSV: Document-level networks • flexible thresholds • networks of documents • scalability issues: “hairballs” • what can we learn about the learners?

  25. Learner Models (Interaction-based) • work with Eric Bruillard, Christophe Reffay and François-Marie Blondel at ENS Cachan • modelling learner communities as the products of interactions among their members

  26. Learner Models (Content-based) • time-consuming and/or intrusive • variety of ITS-based approaches • ours is simpler:

  27. You are what you write! • our approach is to use LSA to create a model of the learner that is based on all of what they have contributed to a community’s discourse space • we have a vector representation for each user • we can then start looking at the relationships between user models

  28. Latent Semantic User Models • can determine similarity between any two user models (again, by calculating cosine) • can start looking at hypotheses about the conditions under which we might expect to see productive interactions (e.g. Vygotsky’s Zone of Proximal Development)

  29. KISSME • Knowledge, • Interaction, and • Semantic • Student • Model • Explorer

  30. Social Network Adjacency Matrix

  31. Semantic Network Adjacency Matrix (high values darkest)

  32. Semantic Network Adjacency Matrix (intermediate values darkest)

  33. Stakeholders • students • teachers • researchers

  34. Students “If you want to take a look at someone who writes stuff that’s really similar to your stuff, check out Charlotte’s contributions. For inspiration, check out Ilsa’s work.”

  35. Teachers “Most of the activity in the class is centered on the topics of sex and drugs. New topics this week include rock and roll. There are three cliques in the class, each with seven students.”

  36. Researchers • visualizations that need some refinement

More Related