Pregled bibliografske jedinice broj: 920520
The Role of Homograms in Machine Translation
The Role of Homograms in Machine Translation // International journal of machine learning and computing (IJMLC), 8 (2018), 2; 90-97 doi:10.18178/ijmlc.2018.8.2.669 (međunarodna recenzija, članak, znanstveni)
CROSBI ID: 920520 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
The Role of Homograms in Machine Translation
Autori
Načinović Prskalo, Lucia ; Brkić Bakarić, Marija
Izvornik
International journal of machine learning and computing (IJMLC) (2010-3700) 8
(2018), 2;
90-97
Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni
Ključne riječi
disambiguation of homograms ; lexicon of homograms ; pitch accent language ; word sense disambiguation
Sažetak
The Croatian language is a pitch-accent language in which the tone contour realized in the stressed syllable carries the lexical information. Therefore, in some cases, different lexical accent gives the word a different meaning. In such cases, the ambiguity of the word in written texts where accents are not usually marked, can be solved by determining the appropriate accent. There are also cases when various basic and derived forms of words have different meanings, different morphosyntactic descriptions and possibly different accents. When words have same written forms, but different meanings, they are called homograms. In order to resolve the ambiguity of homograms, we created a lexicon of homograms which is comprised of all Croatian nouns of different gender which have the same written forms (if accents are not marked), but different meanings, morphosyntactic description and possibly different accents. The lexicon consists of 19, 366 entries and 3, 460 unique homograms. Each entry in the lexicon comprises of the homogram (unaccented word), the accented word, the corresponding MSD, and the accented lemma. The obtained lexicon enables us to identificate and disambiguate homograms within the corpus efficiently and accurately. We also evaluated and analysed the performance of MT systems for the CroatianEnglish language pair with the special emphasis on homogram translation. We confirmed that the disambiguation of homograms can improve the performance of MT systems in avoiding major translation mistakes related to assigning the wrong meaning to homograms
Izvorni jezik
Engleski
Znanstvena područja
Informacijske i komunikacijske znanosti
POVEZANOST RADA
Projekti:
13.13.1.3.03
16.13.2.2.01
Ustanove:
Fakultet informatike i digitalnih tehnologija, Rijeka