Three Approaches to Generating Texts in Different Styles

Ehud Baruch Reiter, Sandra Williams

Research output: Chapter in Book/Report/Conference proceedingPublished conference contribution

1 Citation (Scopus)

Abstract

Natural Language Generation (NLG) systems generate texts in English and other human languages from non-linguistic input data. Usually there are a large number of possible texts that can communicate the input data, and NLG systems must choose one of these. We argue that style can be used by NLG systems to choose between possible texts, and explore how this can be done by (1) imitating a genre style, and (3) imitating an individual's style.
Original languageEnglish
Title of host publicationStyle in Text: Creative Generation and Identification of Authorship
Subtitle of host publicationAISB 2008 Proceedings Volume 7
PublisherAISB
Pages26-33
Number of pages8
Volume7
ISBN (Print) 1 902956 66 4
Publication statusPublished - 2008
EventAISB'08 Symposium Style in Text: Creative Generation and Identification of Authorship - University of Aberdeen, Aberdeen, United Kingdom
Duration: 3 Apr 20083 Apr 2008
http://www.aisb.org.uk/convention/aisb08/index.html

Conference

ConferenceAISB'08 Symposium Style in Text
Country/TerritoryUnited Kingdom
CityAberdeen
Period3/04/083/04/08
Internet address

Bibliographical note

We would like to thank our colleagues in Aberdeen and Milton Keynes, the anonymous reviewers, and the tutors we worked with in skillsum for their insightful comments and suggestions. This work was funded by PACCIT-LINK grant ESRC RES-328-25-0026.

Cite this