A Scalable Approach for Computing Semantic Relatedness using Semantic Web Data

Dennis Diefenbach (1) , Ricardo Usbeck (2) , Kamal Singh (1) , Pierre Maret (1)

1 LHC - Laboratoire Hubert Curien
2 Liepzig university

Computing semantic relatedness is an essential operation for many natural language processing (NLP) tasks, such as Entity Linking (EL) and Question Answering (QA). It is still challenging to find a scalable approach to compute the semantic relatedness using Semantic Web data. Hence, we present for the first time an approach to pre-compute the semantic relatedness between the instances, relations, and classes of an ontology, such that they can be used in real-time applications.

Mots clés: question answering; semantic web; Entity Linking; semantic relatedness

Domaines: Intelligence artificielle [cs.AI]; Base de données [cs.DB]; Informatique et langage [cs.CL]; Web

Dennis Diefenbach, Ricardo Usbeck, Kamal Singh, Pierre Maret. A Scalable Approach for Computing Semantic Relatedness using Semantic Web Data. Proceedings of the 6th International Conference on Web Intelligence, Mining and Semantics, Jun 2016, Nimes, France. ⟨10.1145/2912845.2912864⟩. ⟨hal-01341205⟩