Querying the fragments of english

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Controlled languages are fragments of natural languages stripped clean of lexical, structural and semantic ambiguity. They have been proposed as a means for providing natural language front-ends to access structured knowledge sources, given that they compositionally and deterministically translate into the (logic-based) formalisms such back-end systems support. An important issue that arises in this context is the semantic data complexity of accessing such information (i.e., the computational complexity of querying measured w.r.t. the number of instances declared in the back-end knowledge base or database). In this paper we study the semantic data complexity of a distinguished family of context-free controlled fragments, viz., Pratt and Third's fragments of English. In doing so, we pinpoint those fragments for which the reasoning problems are tractable (in PTime) or intractable (NP-hard or coNP-hard).

Original languageEnglish
Title of host publicationLogic, Language, Information and Computation - 18th International Workshop, WoLLIC 2011, Proceedings
PublisherSpringer Verlag
Pages257-271
Number of pages15
ISBN (Print)9783642209192
DOIs
StatePublished - 2011
Externally publishedYes

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6642 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Keywords

  • computational semantics
  • Controlled language interfaces
  • knowledge base query answering and satisfiability
  • resolution proof procedures
  • semantic data complexity

Fingerprint

Dive into the research topics of 'Querying the fragments of english'. Together they form a unique fingerprint.

Cite this