Document Open Access Logo

Dictionary Alignment by Rewrite-based Entry Translation

Authors Alberto Simões, Xavier Gómez Guinovart



PDF
Thumbnail PDF

File

OASIcs.SLATE.2013.237.pdf
  • Filesize: 431 kB
  • 11 pages

Document Identifiers

Author Details

Alberto Simões
Xavier Gómez Guinovart

Cite AsGet BibTex

Alberto Simões and Xavier Gómez Guinovart. Dictionary Alignment by Rewrite-based Entry Translation. In 2nd Symposium on Languages, Applications and Technologies. Open Access Series in Informatics (OASIcs), Volume 29, pp. 237-247, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2013)
https://doi.org/10.4230/OASIcs.SLATE.2013.237

Abstract

In this document we describe the process of aligning two standard monolingual dictionaries: a Portuguese language dictionary and a Galician synonym dictionary. The main goal of the project is to provide an online dictionary that can show, in parallel, definitions and synonyms in Portuguese and Galician for a specific word, written in Portuguese or Galician. These two languages are very close to each other, and that is the main reason we expect this idea to be viable. The main drawback is the lack of a good and free translation dictionary between these two languages, namely, a dictionary that can cover lexicons with more than one hundred thousand different words. To solve this issue we defined a translation function, based on substitutions, that is able to achieve an F_1 score of 0.88 on a manually verified dictionary of nine thousand words. Using this same translation function to align a Portuguese--Galician dictionary we obtained almost 50% of the dictionary lexicon (more than eighty thousand words) alignment.
Keywords
  • translation
  • rewrite system
  • dictionary
  • dictionary alignment

Metrics

  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    0
    PDF Downloads
Questions / Remarks / Feedback
X

Feedback for Dagstuhl Publishing


Thanks for your feedback!

Feedback submitted

Could not send message

Please try again later or send an E-mail