Using domain ontologies for finding experts in corporate wikis

  • Authors:
  • Ralph Schäfermeier;Adrian Paschke

  • Affiliations:
  • Freie Universität Berlin, Germany;Freie Universität Berlin, Germany

  • Venue:
  • Proceedings of the 7th International Conference on Semantic Systems
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Finding experts is a relevant problem in large, distributed organizations, and automated solutions are needed. In this paper, we propose an approach for finding experts among Wiki authors, since Wikis have emerged as important collaboration and knowledge management tool in enterprizes. By analyzing revision histories and by semantically mapping Wiki contributions to concepts defined in corporate domain ontologies we identify experts. We apply semantic similarity metrics in order to detect references to ontology topics not explicitly mentioned in the text. Furthermore, we use information from the revision history in order to assess the level of expertise and examine the collaborative peer-reviewing processes happening in Wiki systems in order to calculate a reputation score for each author, based on the author's contribution lifetime. We evaluated our approach on the Eclipse project Wiki and conducted a survey with Eclipse project members to assess the quality of our expert finding approach. The results show that the approach yields accurate expertise information.