Williams, Sandra and Reiter, Ehud
Deriving content selection rules from a corpus of non-naturally occurring documents for a novel NLG application.
In: Proceedings of the Corpus Linguistics 2005 Workshop on Using Corpora for Natural Language Generation, 14 July 2005, Birmingham, UK, Information Technology Research Institute (ITRI), pp. 41–48.
Full text available as:
We describe a methodology for deriving content selection rules for NLG applications that aim to replace oral communications from human experts by written communications that are generated automatically. We argue for greater involvement of users and for a strategy for handling sparse data.
Actions (login may be required)