Pivot-based hybrid machine translation to support multilingual communication for closely related languages

Nasution, Arbi Haza (2018) Pivot-based hybrid machine translation to support multilingual communication for closely related languages. World Transactions on Engineering and Technology Education, 16 (2).

Text J4.pdf - Published Version Download (96kB)

Abstract

Machine translation (MT) is very useful in supporting multicultural communication. Existing statistical machine translation (SMT), which requires high quality and quantity of corpora, and rule-based machine translation (RBMT), which requires bilingual dictionaries, morphological, syntax and semantic analysers, are scarce for low-resource languages. Due to the lack of language resources, it is difficult to create MT from high-resource languages to low-resource languages, such as Indonesian ethnic languages. Nevertheless, due to Indonesian ethnic languages’ characteristics, a pivot-based hybrid machine translation (PHMT) can be introduced by combining SMT and RBMT with Indonesian as a pivot, which then can be utilised in a multilingual communication support system. The PHMT translation quality was evaluated, with fluency and adequacy as metrics, and then the usability of the system was evaluated. Despite the medium average translation quality (3.05 fluency score and 3.06 adequacy score), the 3.71 average mean score of the usability evaluation indicates that the system is usable to support multilingual collaboration.

Actions (login required)

View Item