Constructing lexicons of Canadian English

Language technology-for example, spelling checking-is widely used. A key
component of many language technology systems is a computational lexicon-a
dictionary-like knowledge source containing a list of words and information associated with each word. Such resources require great human effort to create and are therefore very expensive. However, language is constantly changing (e.g., new words come into use), and there may be dialectal differences within a language. Therefore, many language technology systems use a computational lexicon which is either out-of-date or inappropriate for the system's target users. The goal of this project is to develop methods for automatically creating and updating computational lexicons that reflect the language used by a particular community, specifically Canadian English. This research will directly address a daunting challenge in language technology which Ontario corporations are struggling to solve, and will facilitate the creation of language technology that is better suited to the needs of Canadians than is currently available.

Faculty Supervisor:

Dr. Graeme Hirst

Student:

Paul Cook

Partner:

Discipline:

Computer science

Sector:

Information and communications technologies

University:

University of Toronto

Program: