1 / 27

Toward Mixed-Initiative Clustering

Toward Mixed-Initiative Clustering. Yifen Huang Tom M. Mitchell Carnegie Mellon University Agents that Learn from Human Teachers March 23, 2009. Unsupervised clustering: A machine builds the model alone. Semi-supervised clustering: A user performs an oracle role. Mixed-Initiative Clustering.

godina
Download Presentation

Toward Mixed-Initiative Clustering

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. Toward Mixed-Initiative Clustering Yifen HuangTom M. MitchellCarnegie Mellon UniversityAgents that Learn from Human TeachersMarch 23, 2009

  2. Unsupervised clustering:A machine builds the model alone. Semi-supervised clustering:A user performs an oracle role.

  3. Mixed-Initiative Clustering Unsupervised clustering:A machine builds the model alone. Semi-supervised clustering:A user performs an oracle role.

  4. Key Question • How can autonomous clustering algorithms be extended to enable mixed-initiative clusteringapproaches involving an iterative sequence of computer-suggested and user-suggested revisionsto converge to a useful hierarchical clustering? • From autonomous clustering to mixed-initiative clustering • From flat feedback to hierarchical feedback

  5. Activity X contains this list of emails: • ## ## ## ## ## An email from Andrea Thomaz belongs to your AAAI symposium activity. Adam Cheyer is a key-person to your CALO activity.

  6. Activity X contains this list of emails: • ## ## ## ## ## • What the hell is this?? DELETE! An email from Andrea Thomaz belongs to your AAAI symposium activity. Too lazy to comment. This is correct. Adam Cheyer is a key-person to your CALO activity.

  7. Framework for Mixed-Initiative Clustering Computer-to-user language: hypotheses User-to-computer language: modified hypotheses Model adaptation algorithm

  8. User Interface

  9. Communicative Languages inSemi-Supervised Clustering Cluster Document

  10. Communicative Languages inSemi-Supervised Clustering ConfirmRemove Cluster Document

  11. Enriching Languages in Flat Clustering ConfirmRemove Cluster Document Word Person

  12. Enriching Languages in Flat Clustering ConfirmRemove ConfirmRemove Cluster Document ConfirmRemove Word Person

  13. Enriching Languages in Hierarchical Clustering ConfirmRemove ConfirmRemove Cluster Cluster Document ConfirmRemove Word Person

  14. Enriching Languages in Hierarchical Clustering ConfirmRemove MoveMergeAddSplit ConfirmRemove Cluster Cluster Document Move ConfirmRemove Move Word Person

  15. Experiment Design • Can mixed-initiative clustering help a user achieve the result faster? • Can mixed-initiative clustering help a machine build a better model?

  16. Dataset • An email dataset of one of the authors • 623 emails • 6684 unique words and 135 individual people • Manually sorted into a hierarchy of 15 cluster nodes including a root, 3 intermediate nodes and 11 leaf nodes

  17. Feedback Sessions • Five initial hierarchical clustering results • Two feedback sessions on each result • Diligent session • Lazy session

  18. Diligent User ConfirmRemove MoveMergeAddSplit ConfirmRemove Cluster Cluster Document Move ConfirmRemove Move Word Person

  19. Lazy User ConfirmRemove MoveMergeAddSplit ConfirmRemove Cluster Cluster Document Move ConfirmRemove Move Word Person

  20. Lazy User vs. Diligent User

  21. Measurement • User feedback is equivalent to edge modification. • Edge Modification Ratio (EMR)equals the ratio of edges needed to be modified in order to reach the reference hierarchy.

  22. Good Results (4/5)

  23. Bad Result (1/5)

  24. One More StepToward Mixed-Initiative Clustering Yifen HuangTom M. MitchellCarnegie Mellon UniversityAgents that Learn from Human TeachersMarch 23, 2009

  25. Low-Latency Mixed-Initiative Clustering

  26. Elaborated Framework for Mixed-Initiative Clustering

  27. Future Work • Feasibility study of the low-latency mixed-initiative interface

More Related