Detail publikace

A New Approach to Pseudoword Generation

Originální název

A New Approach to Pseudoword Generation

Anglický název

A New Approach to Pseudoword Generation

Jazyk

en

Originální abstrakt

Sense-tagged corpora are used to evaluate word sense disambiguation (WSD)systems. Manual creation of such resources is often prohibitively expensive.That is why the concept of pseudowords - conflations of two or moreunambiguous words - has been integrated into WSD evaluation experiments. Thispaper presents a new method of pseudoword generation which takes into accountsemantic-relatedness of the candidate words forming parts of the pseudowords tothe particular senses of the word to be disambiguated. We compare the newapproach to its alternatives and show that the results on pseudowords, that aremore similar to real ambiguous words, better correspond to the actual results.Two techniques assessing the similarity are studied - the first one takesadvantage of manually created dictionaries (wordnets), the second one builds onthe automatically computed statistical data obtained from large corpora. Prosand cons of the two techniques are discussed and the results on a  standard taskare demonstrated.

Anglický abstrakt

Sense-tagged corpora are used to evaluate word sense disambiguation (WSD)systems. Manual creation of such resources is often prohibitively expensive.That is why the concept of pseudowords - conflations of two or moreunambiguous words - has been integrated into WSD evaluation experiments. Thispaper presents a new method of pseudoword generation which takes into accountsemantic-relatedness of the candidate words forming parts of the pseudowords tothe particular senses of the word to be disambiguated. We compare the newapproach to its alternatives and show that the results on pseudowords, that aremore similar to real ambiguous words, better correspond to the actual results.Two techniques assessing the similarity are studied - the first one takesadvantage of manually created dictionaries (wordnets), the second one builds onthe automatically computed statistical data obtained from large corpora. Prosand cons of the two techniques are discussed and the results on a  standard taskare demonstrated.

BibTex


@inproceedings{BUT35423,
  author="Lubomír {Otrusina} and Pavel {Smrž}",
  title="A New Approach to Pseudoword Generation",
  annote="Sense-tagged corpora are used to evaluate word sense disambiguation (WSD)systems.
Manual creation of such resources is often prohibitively expensive.That is why
the concept of pseudowords - conflations of two or moreunambiguous words - has
been integrated into WSD evaluation experiments. Thispaper presents a new method
of pseudoword generation which takes into accountsemantic-relatedness of the
candidate words forming parts of the pseudowords tothe particular senses of the
word to be disambiguated. We compare the newapproach to its alternatives and show
that the results on pseudowords, that aremore similar to real ambiguous words,
better correspond to the actual results.Two techniques assessing the similarity
are studied - the first one takesadvantage of manually created dictionaries
(wordnets), the second one builds onthe automatically computed statistical data
obtained from large corpora. Prosand cons of the two techniques are discussed and
the results on a  standard taskare demonstrated.",
  address="European Language Resources Association",
  booktitle="Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)",
  chapter="35423",
  edition="NEUVEDEN",
  howpublished="print",
  institution="European Language Resources Association",
  year="2010",
  month="april",
  pages="1--5",
  publisher="European Language Resources Association",
  type="conference paper"
}