Parallel, or Comparable? That Is the Question : The Comparison of Parallel and Comparable Data-based Methods for Bilingual Lexicon Induction

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F22%3A00127476" target="_blank" >RIV/00216224:14330/22:00127476 - isvavai.cz</a>
Result on the web
<a href="https://nlp.fi.muni.cz/raslan/2022/paper8.pdf" target="_blank" >https://nlp.fi.muni.cz/raslan/2022/paper8.pdf</a>
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Parallel, or Comparable? That Is the Question : The Comparison of Parallel and Comparable Data-based Methods for Bilingual Lexicon Induction
Original language description
Extracting translation equivalents from parallel data has been considered the main and most efficient method in the lexicography field. However, parallel data are not always available or sufficient, especially for rare and low-resource language pairs. Translation equivalents obtained from comparable data offer a solution for this problem. This paper compares the performance of some methods that utilize either parallel or comparable data and demonstrates the results on the Estonian-Slovak language combination. We show that comparable data usage aspires to be a viable alternative for low-resource languages or rare language pairs, and we propose a new equation for more effective translation equivalents’ induction.
Czech name
—
Czech description
—

Project
<a href="/en/project/EF19_073%2F0016943" target="_blank" >EF19_073/0016943: Internal grant agency of Masaryk University</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year
2022
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Proceedings of the Sixteenth Workshop on Recent Advances in Slavonic Natural Languages Processing, RASLAN 2022
ISBN
9788026317524
ISSN
2336-4289
e-ISSN
—
Number of pages
10
Pages from-to
4-13
Publisher name
Tribun EU
Place of publication
Brno
Event location
Brno
Event date
Jan 1, 2022
Type of event by nationality
CST - Celostátní akce
UT code for WoS article
—

Similar results(10)