Integrating corpus-based and rule-based approaches in an open-source machine translation system

This document discusses integrating corpus-based and rule-based approaches in the open-source machine translation system Apertium. It presents Apertium, an open-source shallow-transfer machine translation engine, and discusses using target language information by incorporating statistical machine translation techniques into Apertium's rule-based architecture. The document was presented at the New Approaches to Machine Translation workshop in Leuven, Belgium on January 11, 2007.