This page last changed on Feb 23, 2008 by martinmueller@northwestern.edu.

Present: Bill Parod (chair and secretary), Phil Burn, John Norstad, Joe Paris, Brian Pytlik Zillig, Amit Kumar, Loretta Auvil, and Martin Mueller

  • Progress reports

Amit: Proxy layer work, new calls have been implemented. Will move users/worksets/projects code onto new Monk workbench with James' help.

John Norstad: Started working on Pryor, which builds data stor from TEI-A files.

Brian: Working on hyphenation issues so that users can have an unadorned version that is TEI P5. DOcumentation on sequence of operations within Abbott. TCP texts are 90% done. Ran Witchcraft texts through Abbott and have forwarded.

Pib: Martin has been working on new training for Early Modern English - major set of correction since last summer. Checking consistencies in the training data. Anticipate changes to spelling maps. This is labor intensive. Expect new release of MorphAdorner with new data next week.

  • Sara Steger Use Case

Obtained preliminary results at the time of the Maryland meeting based on training set at that time. Expect to do NB based on new training data. Amit has done more work on this. It will invoke flows that can be called from the workbench.

  • Should analytics and data cells be merged?

Analytics needs to be in constant contact with data stor so analytic requirements result in actionable specifications.

Focused groups on specific tasks is more productive.

Some groups around specific tasks change as the project goes on. Use existing call times for analytics and data cell but schedule small groups on specific topics as needed.

Martin: If Sara's use case is developed, can it be done in such a way that other statistical routines can be plugged in?

Amit: We have the NB routines in hand. Once we have set up in the Monk environment to work with those, we can substitute other modules. Clustering modules, for example, will likely need different visualizations / interfaces. This need for other visualization techniques might be a motivation to integrate ManyEyes.

  • New Chair
    Martin will chair the combined data and analytics cell

New data cell chairProgress reports

Document generated by Confluence on Apr 19, 2009 15:04