We present an overview of Amalgam, a sentence realization module that combines machine-learned and knowledge-engineered components to produce natural language sentences from logical form inputs. We describe the decomposition of the task of sentence realization into a linguistically-informed series of steps, with particular attention to the linguistic issues that arise in German. We report on the evaluation of component steps and of the overall system.

PDF file

Publisher: Association for Computational Linguistics
All copyrights reserved by ACL 2002.

Details