This paper describes Generation-Heavy Hybrid Machine Translation (GHMT), a novel approach for trans lating between structurally-divergent language pairs with asymmetrical resources. The approach depends on the existence of rich target language resources such as word lexical semantics, categorial variations and subcategorization frames. These resources are used to overgenerate multiple lexico-structural variations from a target-glossed syntactic dependency representation of the source language sentence. This symbolic overgeneration, which accounts for a wide range of possible variations, is constrained by a statistical target language model. The exploitation of target language resources (symbolic and statistical) to handle a problem usually reserved for Transfer and Interlingual MT is useful for translation from source languages with scarce linguistic resources. A preliminary evaluation on the application of this approach to Spanish-English MT is conducted with promising results.
|Original language||English (US)|
|State||Published - 2002|
ASJC Scopus subject areas
- Language and Linguistics