Semantic Web Dog Food

Presented at: 4th European Semantic Web Conference (ESWC2007)

by Norberto Fernandez Garcia, Jose Maria Blazquez del Toro, Luis Sanchez Fernandez, Ansgar Bernardi

In this paper we introduce the IdentityRank algorithm, developed as part of the EU-funded project NEWS to address the problem of named entity disambiguation in the context of semantic annotation of ne ws items. The algorithm provides a ranking of the candidate instances within an ontology which can be associated to a certain entity. In order to do so, it uses as context the metadata available in a certain news item. The algorithm has been evaluated with promising results.

Keywords: Entity disambiguation, Semantic Annotation

Resource URI on the dog food server: http://data.semanticweb.org/conference/eswc/2007/paper-27

Explore this resource elsewhere:

• on Snorql
• on Sindice
• on dotAC Explorer