Publication detail

Deep Learning from Web-Scale Corpora for Better Dictionary Interfaces

OTRUSINA, L. SMRŽ, P.

Original Title

Deep Learning from Web-Scale Corpora for Better Dictionary Interfaces

Type

article in a collection out of WoS and Scopus

Language

English

Original Abstract

This paper explores advanced learning mechanisms - neural networks trained by the Word2Vec method - for predicting word associations. We discuss how the approach can be built into dictionary interfaces to help tip-of-the-tongue searches. We also describe our contribution to the CogALex 2014 shared task. We argue that the reverse response-stimulus word associations chosen for the shared task are only mildly related to the motivation idea of the lexical access support system. The methods employed in our contribution are briefly introduced. We present results of experiments with various parameter settings and show what improvement can be expected if more than one answer is allowed. The paper concludes with a proposal for a new collective effort to assemble real tip-of-the-tongue situation records for future, more-realistic evaluations.

Keywords

Word2Vec, neural networks, ClueWeb, UKWaC, tip-of-the-tongue phenomenon

Authors

OTRUSINA, L.; SMRŽ, P.

RIV year

2014

Released

31. 7. 2014

Publisher

Association for Computational Linguistics

Location

Dublin

ISBN

978-1-63439-217-4

Book

Proceedings of the 4th Workshop on Cognitive Aspects of the Lexicon (CogALex)

Pages from

22

Pages to

30

Pages count

9

URL

http://www.aclweb.org/anthology/W/W14/W14-4703.pdf

BibTex

@inproceedings{BUT111643,
  author="Lubomír {Otrusina} and Pavel {Smrž}",
  title="Deep Learning from Web-Scale Corpora for Better Dictionary Interfaces",
  booktitle="Proceedings of the 4th Workshop on Cognitive Aspects of the Lexicon (CogALex)",
  year="2014",
  pages="22--30",
  publisher="Association for Computational Linguistics",
  address="Dublin",
  isbn="978-1-63439-217-4",
  url="http://www.aclweb.org/anthology/W/W14/W14-4703.pdf"
}