wiki:PimoContentTraining

PimoContentTraining

The [catwiesel engine] developed by DFKI is able to suggest which concept from the PIMO may be suited for a piece of information. The PimoContentTraining system has the task to suggest these topics (or new topics, if needed) to the user anytime the user approaches documents that are not related to the PIMO already.

The PimoContentTraining system is trained before it can work. Either by linking folders to pim classes/instances or by adding single documents to the class/instance. The catwiesel engine is based on a lucene index and learns the mental models of the user.

This was already published by us several times:

  • "Leveraging Passive Paper Piles to Active Objects in Personal Knowledge Spaces. Heiko Maus, Harald Holz, Ansgar Bernardi, Oleg Rostanin" draft
  • "Overview and Outlook on the Semantic Desktop, Leo Sauermann, Ansgar Bernardi, A. Dengel" paper
Last modified 11 years ago Last modified on 02/12/06 23:25:48