• Documents
  • Authors
  • Tables
  • Log in
  • Sign up
  • MetaCart
  • DMCA
  • Donate

CiteSeerX logo

Advanced Search Include Citations
Advanced Search Include Citations

DMCA

Deterministic Part-of-Speech Tagging with Finite-State Transducers (1995)

Cached

  • Download as a PDF

Download Links

  • [www.cs.mun.ca]
  • [web.cs.mun.ca]
  • [www.merl.com]
  • [www.merl.com]
  • [www.merl.com]
  • [www.merl.com]
  • [www.merl.com]
  • [acl.ldc.upenn.edu]
  • [www.aclweb.org]
  • [www.aclweb.org]
  • [ucrel.lancs.ac.uk]
  • [wing.comp.nus.edu.sg]
  • [aclweb.org]
  • [aclweb.org]
  • [www.dei.unipd.it]
  • [www.dei.unipd.it]
  • [wing.comp.nus.edu.sg]

  • Other Repositories/Bibliography

  • DBLP
  • Save to List
  • Add to Collection
  • Correct Errors
  • Monitor Changes
by Emmanuel Roche , Yves Schabes
Venue:Computational Linguistics
Citations:96 - 0 self
  • Summary
  • Citations
  • Active Bibliography
  • Co-citation
  • Clustered Documents
  • Version History

BibTeX

@ARTICLE{Roche95deterministicpart-of-speech,
    author = {Emmanuel Roche and Yves Schabes},
    title = {Deterministic Part-of-Speech Tagging with Finite-State Transducers},
    journal = {Computational Linguistics},
    year = {1995},
    volume = {21},
    pages = {227--253}
}

Share

Facebook Twitter Reddit Bibsonomy

OpenURL

 

Abstract

Stochastic approaches to natural language processing have often been preferred to rule-based approaches because of their robustness and their automatic training capabilities. This was the case for part-of-speech tagging until Brill showed how state-of-the-art part-of-speech tagging can be achieved with a rule-based tagger by inferring rules from a training corpus. However, current implementations of the rule-based tagger run more slowly than previous approaches. In this paper, we present a finite-state tagger, inspired by the rule-based tagger, that operates in optimal time in the sense that the time to assign tags to a sentence corresponds to the time required to follow a single path in a deterministic finite-state machine. This result is achieved by encoding the application of the rules found in the tagger as a nondeterministic finite-state transducer and then turning it into a deterministic transducer. The resulting deterministic transducer yields a part-of-speech tagger whose speed is dominated by the access time of mass storage devices. We then generalize the techniques to the class of transformation-based systems. 1.

Keyphrases

deterministic part-of-speech tagging    rule-based tagger    finite-state transducer    optimal time    nondeterministic finite-state transducer    state-of-the-art part-of-speech tagging    deterministic transducer    single path    deterministic finite-state machine    part-of-speech tagging    mass storage device    natural language processing    transformation-based system    stochastic approach    rule-based approach    automatic training capability    current implementation    access time    deterministic transducer yield    previous approach    part-of-speech tagger    finite-state tagger    training corpus   

Powered by: Apache Solr
  • About CiteSeerX
  • Submit and Index Documents
  • Privacy Policy
  • Help
  • Data
  • Source
  • Contact Us

Developed at and hosted by The College of Information Sciences and Technology

© 2007-2019 The Pennsylvania State University