Refer-to-as Relations as Semantic Knowledge


  • Song Feng IBM T.J. Watson Research Center / Stony Brook University
  • Sujith Ravi Google
  • Ravi Kumar Google
  • Polina Kuznetsova Stony Brook University
  • Wei Liu University of North Carolina at Chapel Hill
  • Alexander Berg University of North Carolina at Chapel Hill
  • Tamara Berg University of North Carolina at Chapel Hill
  • Yejin Choi University of Washington



We study Refer-to-as relations as a new type of semanticknowledge. Compared to the much studied Is-a relation,which concerns factual taxonomy knowledge, Refer-to-as relationsaim to address pragmatic semantic knowledge. Forexample, a “penguin” is a “bird” from a taxonomy point ofview, but people rarely refer to a “penguin” as a “bird” invernacular use. This observation closely relates to the entrylevelcategorization studied in Prototype Theory in Psychology.We posit that Refer-to-as relations can be learned fromdata, and that both textual and visual information would behelpful in inferring the relations. By integrating existing lexicalstructure knowledge with language statistics and visualsimilarities, we formulate a collective inference approach tomap all object names in an encyclopedia to commonly usednames for each object. Our contributions include a new labeleddata set, the inference and optimization approach, andthe computed mappings and similarities.




How to Cite

Feng, S., Ravi, S., Kumar, R., Kuznetsova, P., Liu, W., Berg, A., Berg, T., & Choi, Y. (2015). Refer-to-as Relations as Semantic Knowledge. Proceedings of the AAAI Conference on Artificial Intelligence, 29(1).



Main Track: NLP and Knowledge Representation