/documentextractor

A web application to process documents into extractions and annotate those extractions.

Primary LanguageCSS

There are a number of hidden URLs defined in the routes file (conf/routes) that provide methods of getting Precision-Yield curves as well as annotation data in different formats.

/logentry/:i/annotate/:name
/logentry/:i/gold/:name
/logentry/:i/sentences
/logentry/:i/py/:name
/logentry/:i/graph/:name

For example, if you extract a document and use the permalink (look for the link named "link to this page") to the extraction page, you will have an id for that document (Let's say it's 5). Now you can visit the page /logentry/5/annotate/michael and annotate that document using the name "michael". Next, visit /logentry/5/py/michael and see the PY points from Michael's annotations.

This project currently uses Open IE 4 and ReVerb, but other Open IE extractors can be easily added for annotations.