Skip to content

Topic Analysis

Jump to bottom
knoxa edited this page May 22, 2017 · 2 revisions

Clustering by topic with Carrot2

The Carrot2 Workbench will cluster text in C2 XML format, or you can supply any XML plus an XSLT stylesheet that will convert it to C2 XML. For example, to cluster the RDF/XML kidnapping events from Extracted Information:

  1. In Carrot2 Workbench, select XML as the source.
  2. Under Basic, specify the XML Resource: http://dstl.github.io/muc3/events/kidnap.rdf
  3. Under Medium, specify the XSLT Stylesheet: http://dstl.github.io/muc3/xsl/events-carrot.xsl
  4. Click the Process button
Clone this wiki locally