The main aim of the CONTROL project is to promote application of constraint solving technology in the field of natural language processing (NLP). The motivation for this line of research is primarily to increase the robustness of language processing systems. Constraint solving provides a framework for linking and interleaving the various phases of language processing, while retaining their conceptual distinction. This allows a much more flexible use of available syntactic, semantic, contextual and pragmatic information in order to understand, generate or learn language. Current research by members of the project has already produced results based on this approach, including resolving ambiguity and improved error-robustness.
Kickoff meeting 17 March 2004
5th International Workshop on Constraints and Language Processing - CSLP 2008
The CONTROL project is supported by SNF