Per our e-mail discussion, I'm making a note here that @ebeshero's next step in developing our Annotation Tool's XQuery will be to refine the code so that the tool first pulls the top three most frequently co-occurring categories appearing together with a particular named item. (Sample categories: people, places, organizations, etc.: see Section B under "Contextual/Relational Markup," within our Coding Guidelines, at http://bit.ly/1PdAKRJ.)
Following that, the XQuery will then select the top three named entities from within each of these main "top three" groups.
At present, the tool's output displays as a chart. Further steps will include exploring possibilities for SVG visualization of the result clusters, which will both aid the writing of site index entries for named entities and may eventually even be placed inline with the entities themselves, when they occur in the archive.
Per our e-mail discussion, I'm making a note here that @ebeshero's next step in developing our Annotation Tool's XQuery will be to refine the code so that the tool first pulls the top three most frequently co-occurring categories appearing together with a particular named item. (Sample categories: people, places, organizations, etc.: see Section B under "Contextual/Relational Markup," within our Coding Guidelines, at http://bit.ly/1PdAKRJ.)
Following that, the XQuery will then select the top three named entities from within each of these main "top three" groups.
At present, the tool's output displays as a chart. Further steps will include exploring possibilities for SVG visualization of the result clusters, which will both aid the writing of site index entries for named entities and may eventually even be placed inline with the entities themselves, when they occur in the archive.