Paper: Chunk Parsing Revisited

ACL ID W05-1514
Title Chunk Parsing Revisited
Venue Workshop On Parsing Technology
Year 2005
  • Yoshimasa Tsuruoka (University of Tokyo, Tokyo Japan; CREST Japan Science and Technology Corporation, Saitama Japan)
  • Jun'ichi Tsujii (University of Tokyo, Tokyo Japan; University of Manchester, Manchester UK; CREST Japan Science and Technology Corporation, Saitama Japan)

Chunk parsing is conceptually appealing but its performance has not been satis- factory for practical use. In this pa- per we show that chunk parsing can perform significantly better than previ- ously reported by using a simple sliding- window method and maximum entropy classifiers for phrase recognition in each level of chunking. Experimental results with the Penn Treebank corpus show that our chunk parser can give high-precision parsing outputs with very high speed (14 msec/sentence). We also present a pars- ing method for searching the best parse by considering the probabilities output by the maximum entropy classifiers, and show that the search method can further im- prove the parsing accuracy.

  author    = {Tsuruoka, Yoshimasa  and  Tsujii, Jun'ichi},
  title     = {Chunk Parsing Revisited},
  booktitle = {Proceedings of the Ninth International Workshop on Parsing Technology},
  month     = {October},
  year      = {2005},
  address   = {Vancouver, British Columbia},
  publisher = {Association for Computational Linguistics},
  pages     = {133--140},
  url       = {}