Skip to main content

Integration of Machine Translation Paradigms

Article Category

Article available in the folowing languages:

Open-source advanced linguistic system

An advanced tool that can efficiently translate language in real time would have an enormous payoff for society. EU-funded scientists proposed an advanced machine translation (MT) paradigm to further enhance the quality of translated texts.

Digital Economy

MT is a highly interdisciplinary and multidisciplinary field requiring input from professionals ranging from translators to engineers to computer scientists to mathematicians to linguists. The EU-funded IMTRAP (Integration of machine translation paradigms) project worked on developing and validating an open-source hybrid MT system. Researchers focused on multiple aspects of linguistics such as morphology, syntax and semantics. The resulting cutting-edge hybrid system prototype combines different MT paradigms, including statistical and rule-based MT (RBMT), and can be trainable in any pair of languages. Researchers successfully introduced baseline statistical MT (SMT) systems for Chinese-to-Spanish and English-to-Spanish through a collection of corpora for these pairs of languages. Another important IMTRAP achievement was development of the first Chinese-to-Spanish open-source hybrid system. The input of this system was pre-processed with an RBMT system and its output was passed to an SMT system. SMT uses models whose parameters stem from the analysis of monolingual and bilingual corpora. RBMT was used to define the structural transfer rules for phrases, and SMT was considered as the only source for the lexical transfer of words. By using SMT techniques, notable enhancements were observed in the final output of translation. Furthermore, the output of this new hybrid system was contrasted with a state-of-the-art SMT system in the out-of-domain test set. Results showed that the new RBMT system outperforms the SMT system in all linguistic levels except the syntax level. Specifically, the new hybrid system far outperformed the state of the art in terms of lexical coverage. In addition, IMTRAP achieved a higher level of hybridisation in statistical and RBMT. Work also focused on extracting transfer rules, assigning a probability to a sequence of n words, and introducing a language model to the generation step. The results of the research, which succeeded in achieving its aim, were published in journal papers and books, as well as through international conferences. Commercialisation of a cost-effective hybrid MT system will have wide-ranging applications in information access systems and document translations. Society at large stands to benefit enormously, as will the European civil service and international relationships, not least with Asian parties since the project focused initially on the Chinese language.


Linguistic, machine translation, IMTRAP, languages, statistical MT

Discover other articles in the same domain of application