Napredna pretraga

Pregled bibliografske jedinice broj: 547340

Automatic and Human Evaluation on English-Croatian Legislative Test Set


Brkić, Marija; Seljan, Sanja; Vičić, Tomislav
Automatic and Human Evaluation on English-Croatian Legislative Test Set // Lecture Notes in Computer Science - LNCS, 7816 (2013), 1; 311-317 doi:10.1007/978-3-642-37247-6 (međunarodna recenzija, članak, znanstveni)


Naslov
Automatic and Human Evaluation on English-Croatian Legislative Test Set

Autori
Brkić, Marija ; Seljan, Sanja ; Vičić, Tomislav

Izvornik
Lecture Notes in Computer Science - LNCS (0302-9743) 7816 (2013), 1; 311-317

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni

Ključne riječi
Automatic evaluation; machine translation; English-Croatia; BLEU; NIST; WER; F-measure
(Automatic evaluation; machine translation; English-Croatia; BLEU; NIST; WER; F-measur)

Sažetak
This paper presents work on the manual and automatic evaluation of the online available machine translation (MT) service Google Translate, for the English-Croatian language pair in legislation and general domains. The experimental study is conducted on the test set of 200 sentences in total. Human evaluation is performed by native speakers, using the criteria of fluency and adequacy, and it is enriched by error analysis. Automatic evaluation is performed on a single reference set by using the following metrics: BLEU, NIST, F-measure and WER. The influence of lowercasing, tokenization and punctuation is discussed. Pearson's correlation between automatic metrics is given, as well as cor relation between the two criteria, fluency and adequacy, and automatic metrics.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti

Napomena
ISBN 978-3-642-37246-9 ; 14th International Conference (CICLing 2013) "Computational Linguistics and Intelligent Text Processing" : proceedings. Part I ; Alexander Gelbukh (ur.) ; Samos, Grčka ; 24.-30.03.2013.



POVEZANOST RADA


Projekt / tema
130-1300646-0909 - Informacijska tehnologija u prevođenju hrvatskoga i e-učenju jezika (Sanja Seljan, )

Ustanove
Filozofski fakultet, Rijeka,
Filozofski fakultet, Zagreb,
Sveučilište u Rijeci - Odjel za informatiku

Časopis indeksira:


  • Web of Science Core Collection (WoSCC)
    • SCI-EXP, SSCI i/ili A&HCI
  • Scopus


Uključenost u ostale bibliografske baze podataka:


  • Compendex (EI Village)
  • MathSciNet
  • Zentrallblatt für Mathematik/Mathematical Abstracts
  • Scopus, ACM - Digital Library, DBLP, Google Scholar, IO-Port, MathSciNeg


Citati