The best of two worlds: cooperation of statistical and rule-based taggers for Czech

  • Authors:
  • Drahomíra "johanka" Spoustová;Jan Hajič;Jan Votrubec;Pavel Krbec;Pavel Květoň

  • Affiliations:
  • Charles University, Prague, Czech Republic;Charles University, Prague, Czech Republic;Charles University, Prague, Czech Republic;Voice Technologies and Systems, Prague, Czech Republic;Academy of Sciences of the Czech Republic

  • Venue:
  • ACL '07 Proceedings of the Workshop on Balto-Slavonic Natural Language Processing: Information Extraction and Enabling Technologies
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Several hybrid disambiguation methods are described which combine the strength of hand-written disambiguation rules and statistical taggers. Three different statistical (HMM, Maximum-Entropy and Averaged Perceptron) taggers are used in a tagging experiment using Prague Dependency Tree-bank. The results of the hybrid systems are better than any other method tried for Czech tagging so far.