An Architecture for Data-to-Text Systems

Research output: Chapter in Book/Report/Conference proceedingConference contribution

82 Citations (Scopus)

Abstract

I present an architecture for data-to-text systems, that is NLG systems which produce texts from non-linguistic input data; this essentially extends the architecture of Reiter and Dale (2000) to systems whose input is raw data instead of AI knowledge bases. This architecture is being used in the BabyTalk project, and is based on experiences in several projects at Aberdeen; it also seems to be compatible with many data-to-text systems developed elsewhere. It consists of four stages which are organised in a pipeline: Signal Analysis, Data Interpretation, Document Planning, and Microplanning and Realisation.
Original languageEnglish
Title of host publicationProceedings of the Eleventh European Workshop on Natural Language Generation (ENLG 07)
EditorsStephan Busemann
Place of PublicationStroudsburg
PublisherAssociation for Computational Linguistics
Pages97-104
Number of pages8
Publication statusPublished - 2007

    Fingerprint

Cite this

Reiter, E. B. (2007). An Architecture for Data-to-Text Systems. In S. Busemann (Ed.), Proceedings of the Eleventh European Workshop on Natural Language Generation (ENLG 07) (pp. 97-104). Stroudsburg: Association for Computational Linguistics.