Machine Translation Using Automatically Inferred Construction-based Correspondence and Language Models?

We discuss the problem of translation in the wider context of the problem of mean- ing in cognition and describe a structural statistical machine translation (MT) method moti- vated by philosophical, cognitive, and computational considerations. Our approach relies on a recently published algorithm capable of learning from a raw corpus a limited yet effec- tive grammar that can be used to construct probabilistic parsers and language models, and on cognitively motivated heuristics for learning construction-based translation models. A pi- lot system has been implemented and tested successfully on simple English to Hebrew and Spanish to English translation tasks.