Trust network-based filtering of aggregated claims

  • Authors:

  • Jennifer Golbeck;Bijan Parsia

  • Affiliations:

  • Maryland Information and Network Dynamics Laboratory, University of Maryland, College Park, MD 20742, USA.;Maryland Information and Network Dynamics Laboratory, University of Maryland, College Park, MD 20742, USA

  • Venue:
  • International Journal of Metadata, Semantics and Ontologies
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

On the semantic web, assertions may be aggregated from many sources, those aggregations filtered, reasoned over, aggregated with other aggregators, displayed, scraped, extracted, recombined, and otherwise processed without significant human oversight. To preserve the connection between assertions and their source, various provenance schemes for semantic web data have been explored. However, the primary focus has been on authenticating the author of a particular statement, e.g., using digital signatures, but there is no provision for relating the authenticity of the source of the assertion and the trustworthiness of the assertion itself. This paper presents a method for using semantic web based trust networks to infer the reputation of sources for a statement and compose the reputation of several sources. By calculating a trust rating for each statement based on the ratings of its sources, the set of statements can be filtered based on the rating.