Fully automatic high quality Machine Translation (MT) is traditionally being considered as one of the most desirable dreams of the mankind, but only limited achievements have been obtained in the past. The interlingual approach is the least studied one and, theoretically, an interlingual representation combined with a thinking machine would be able to translate texts as good as people do. The key idea of this book is to elaborate a limited interlingual MT system that is able to translate subsets between two languages - English and German - with the most adequate translation results. For this purpose, a language-independent Interlingua largely based on the MultiNet knowledge representation system was designed. The created system is able to translate sentences based on a limited vocabulary, to attach logically objects and subjects to the adjectives and to one another.
Index Terms
- Machine Translation Interlingua based on MultiNet
Recommendations
Interlingua-based English–Hindi Machine Translation and Language Divergence
Interlingua and transfer-based approaches to machine translation have long been in use in competing and complementary ways. The former proves economical in situations where translation among multiple languages is involved, and can be used as a knowledge-...
N-gram-based statistical machine translation versus syntax augmented machine translation: comparison and system combination
EACL '09: Proceedings of the 12th Conference of the European Chapter of the Association for Computational LinguisticsIn this paper we compare and contrast two approaches to Machine Translation (MT): the CMU-UKA Syntax Augmented Machine Translation system (SAMT) and UPC-TALP N-gram-based Statistical Machine Translation (SMT). SAMT is a hierarchical syntax-driven ...