Abstract
Two machine translation (MT) systems, a statistical MT (SMT) system and a hybrid system (rule-based and SMT) were tested in order to compare various MT performances. The source language was English (EN) and the target language Portuguese (PT). The SMT tool gave much fewer errors than the hybrid system. Major problem areas of both systems concerned the transfer of verb systems from source to target language, and of the hybrid system the word-to-word translation, since its resources are mainly dictionaries and not corpora.
Highlights
Machine Translation (MT) became one of the first real enterprises to computationally process human language, even before the term computational linguistics was coined
That was the birth of MT and automatically the birth of RuleBased Machine Translation (RBMT)
In order to evaluate the performance of different MT systems, we choose criteria based on the five2 distinct levels of language knowledge according to Akmajian et al (2001): a) Morphology: Morphology includes the correct marking of plurals, case marking, anaphoric pronominalizations, in special the well formation of verbal systems, including right conjugations, right aspect choice, tense choice etc
Summary
Machine Translation (MT) became one of the first real enterprises to computationally process human language, even before the term computational linguistics was coined. Computational linguists attempted to improve system rules based on newer paradigms of linguistic theory. This technique became a new approach around the 1980's when Makoto Nagao proposed in Japan to look at bigger chunks of input in the source language that could be translated into the target language, if bilingual examples of those chunks were already at hand in a certain database (Nagao, 1984). Research in EBMT began and researchers awoke to the possibility of using the output of ongoing research in corpus linguistics by using bilingual corpora as databases for EBMT systems
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
