hrcak mascot   Srce   HID

Izvorni znanstveni članak
https://doi.org/10.2498/cit.1001917

Statistical Machine Translation of Croatian Weather Forecasts: How Much Data Do We Need?

Nikola Ljubešić ; Department of Information Sciences, Faculty of Humanities and Social Sciences, University of Zagreb, Croatia
Petra Bago ; Department of Information Sciences, Faculty of Humanities and Social Sciences, University of Zagreb, Croatia
Damir Boras ; Department of Information Sciences, Faculty of Humanities and Social Sciences, University of Zagreb, Croatia

Puni tekst: engleski, pdf (165 KB) str. 303-308 preuzimanja: 256* citiraj
APA 6th Edition
Ljubešić, N., Bago, P. i Boras, D. (2010). Statistical Machine Translation of Croatian Weather Forecasts: How Much Data Do We Need?. Journal of computing and information technology, 18 (4), 303-308. https://doi.org/10.2498/cit.1001917
MLA 8th Edition
Ljubešić, Nikola, et al. "Statistical Machine Translation of Croatian Weather Forecasts: How Much Data Do We Need?." Journal of computing and information technology, vol. 18, br. 4, 2010, str. 303-308. https://doi.org/10.2498/cit.1001917. Citirano 26.06.2019.
Chicago 17th Edition
Ljubešić, Nikola, Petra Bago i Damir Boras. "Statistical Machine Translation of Croatian Weather Forecasts: How Much Data Do We Need?." Journal of computing and information technology 18, br. 4 (2010): 303-308. https://doi.org/10.2498/cit.1001917
Harvard
Ljubešić, N., Bago, P., i Boras, D. (2010). 'Statistical Machine Translation of Croatian Weather Forecasts: How Much Data Do We Need?', Journal of computing and information technology, 18(4), str. 303-308. doi: https://doi.org/10.2498/cit.1001917
Vancouver
Ljubešić N, Bago P, Boras D. Statistical Machine Translation of Croatian Weather Forecasts: How Much Data Do We Need?. Journal of computing and information technology [Internet]. 2010 [pristupljeno 26.06.2019.];18(4):303-308. doi: https://doi.org/10.2498/cit.1001917
IEEE
N. Ljubešić, P. Bago i D. Boras, "Statistical Machine Translation of Croatian Weather Forecasts: How Much Data Do We Need?", Journal of computing and information technology, vol.18, br. 4, str. 303-308, 2010. [Online]. doi: https://doi.org/10.2498/cit.1001917

Sažetak
This research is the first step towards developing a system
for translating Croatian weather forecasts into multiple
languages. This step deals with the Croatian-English
language pair. The parallel corpus consists of a one-year
sample of the weather forecasts for the Adriatic, consisting
of 7,893 sentence pairs. Evaluation is performed
by the automatic evaluation measures BLUE, NIST and
METEOR, as well as by manually evaluating a sample of
200 translations. We have shown that with a small-sized
training set and the state-of-the artMoses system, decoding
can be done with 96% accuracy concerning adequacy
and fluency. Additional improvement is expected by
increasing the training set size. Finally, the correlation
of the recorded evaluation measures is explored.

Ključne riječi
statistical machine translation; automatic evaluation; manual evaluation; correlation between evaluation measures

Hrčak ID: 63862

URI
https://hrcak.srce.hr/63862

Posjeta: 413 *