@unhammer

Comparing a Linguistic and a Stochastic Tagger

, and . Proceedings of the 8th conference of the EACL, page 246--253. Morristown, NJ, USA, ACL, (1997)
DOI: http://dx.doi.org/10.3115/979617.979649

Abstract

Concerning different approaches to automatic PoS tagging: EngCG-2, a constraint-based morphological tagger, is compared in a double-blind test with a state-of-the-art statistical tagger on a common disambiguation task using a common tag set. The experiments show that for the same amount of remaining ambiguity, the error rate of the statistical tagger is one order of magnitude greater than that of the rule-based one. The two related issues of priming effects compromising the results and disagreement between human annotators are also addressed.

Description

Fantastic. A double-blind experiment to test the accuracy of a tagger believed too good to be true.

Links and resources

Tags

community

  • @lepsky
  • @unhammer
  • @dblp
@unhammer's tags highlighted