Professor of Computational Linguistics

University of Toronto, Department of Computer Science


Syntactic nuances of style and meaning

The nuances of denotation and connotation that are a part of everyday language are a serious problem in many applications of computational linguistics. In machine translation, for example, if a translation of a document (such as a business letter or memo) is to be faithful to the original, it must capture its style as well as its meaning, whether it be friendly, threatening, formal, informal, or even deliberately obscure. Chrysanne DiMarco and Graeme Hirst investigated how a machine translation system can be sensitive to nuances of syntactic structure in the input and preserve their effect in the output, even if the target language requires a completely different structure to achieve the same nuance. This work is continued in our project on lexical nuances of style and meaning.


DiMarco, Chrysanne and Hirst, Graeme. “A computational theory of goal-directed style in syntax.” Computational Linguistics, 19(3), September 1993, 451-499. [PDF]