Learning Verbal Transitivity Using LogLinear Models

  title={Learning Verbal Transitivity Using LogLinear Models},
  author={Nuno Miguel Marques and Jos{\'e} Gabriel Pereira Lopes and Carlos Agra Coelho},
Portuguese dictionaries lack information about the kind of phrases or clauses a verb goes with. In this paper we describe how this information can be computationally learned from an automatically tagged corpus with almost 10,000,000 words. Loglinear modeling for categorical data will be used to analyze and automatically identify the subcatego-rization dependencies. Loglinear models were also used in an unsuper-vised clustering algorithm to accurately determine the verbal transitive-ness… CONTINUE READING

From This Paper

Topics from this paper.
7 Citations
11 References
Similar Papers


Publications referenced by this paper.
Showing 1-10 of 11 references

Reconhecimento deneologismos

  • Nuno C. Marques, Jos e Gabriel Lopes.
  • 1994

Brent . From grammar to lexicon : Unsupervised learning oflexical syntax

  • R. Michael
  • Computacional Linguistics
  • 1993

A practical partofspeech tagger

  • Doug Cutting, Julian Kupiek, Penelope Sibun.
  • 1992

Brent . Automatic acquisition of subcategorization frames fromuntagged text

  • R. Michael
  • 1991

GLIM: An Introduction

  • M.J.R. Healy
  • 1988

A neural network approach topartofspeech tagging

  • Nuno C. Marques, Jos e Gabriel Lopes

Similar Papers

Loading similar papers…