Let's talk about NLP and automated reasoning without getting grey hair

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

Progress in the field of knowledge-based natural language processing is closely related to progress in the field of automated reasoning. Instead of focusing on full natural languages that are probably the most expressive and flexible knowledge representation languages that exist, I will argue that it is a better strategy to look at well-defined subsets of natural language that have the same expressivity as the formal languages that automated theorem provers can process. In particular, I will look at the two description logic reasoners CEL and RacerPro and the first-order logic theorem prover and model-generator EKRHyper and discuss what subsets of natural language can be processed and what kind of inferences are supported by these tools.
Original languageEnglish
Title of host publicationSearching answers
Subtitle of host publicationFestschrift in honour of Michael Hess on the occasion of his 60th birthday
EditorsSimon Clematide, Manfred Klenner, Martin Volk
Place of PublicationMunster, Germany
PublisherVerlagshaus Monsenstein und Vannerdat
Pages119-126
Number of pages8
ISBN (Print)9783865829559
Publication statusPublished - 2009

Publication series

NameMV Wissenschaft
PublisherVerlagshaus Monsenstein und Vannerdat

Keywords

  • knowledge representation
  • description logics
  • first-order logic
  • automated reasoning
  • (controlled) natural language

Fingerprint

Dive into the research topics of 'Let's talk about NLP and automated reasoning without getting grey hair'. Together they form a unique fingerprint.

Cite this