This is demo of Frictionary, a system for extracting and aggregating problems from help requests, bug reports, discussion boards, and any other textual source of software problem discussions. This demo shows data help requests, bug reports, and other types of discussion board posts about Firefox based on ~103,000 posts between November 2009 and July 2010, from support.mozilla.org, superuser.com, Firefox forums on mozillaZine, and Mozilla bug reports. Frictionary uses natural language processing and information extraction techniques to identify problem topics in each post. Questions? Consult the research article on Frictionarya> or contact Andrew J. Ko, Assistant Professor at the University of Washington Information School.

  

sources

Loading feature/action data...