An Approach for Contextual Translation based on Semantic Schemas

Processing natural language statements to obtain equivalent translations in a different language has long been an area of research in Artificial Intelligence. Still, most existing approaches are word-to-word translations and thus cannot ensure high-quality translations by considering the whole meaning of the statement. The translation mechanism presented in this paper constructs semantic schema representations of the sentence syntactic structure and then evaluates them with corresponding constructions in the translation language. Contextual translation is ensured by the semantic schema representations usage that pays attention to the dependency relations existing between the sentence words. Key-Words: Semantic schema, Natural language, Machine translation, Contextual translation