Task-Based Evaluation of NLG Systems: Control vs Real-World Context

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

Abstract

Currently there is little agreement about, or even discussion of, methodologies for taskbased evaluation of NLG systems. I discuss one specific issue in this area, namely the importance of control vs the importance of ecological validity (real-world context), and suggest that perhaps we need to put more emphasis on ecological validity in NLG evaluations.
Original languageEnglish
Title of host publicationProceedings of UCNLG+Eval 2011
PublisherAssociation for Computational Linguistics
Pages28-32
Number of pages5
Publication statusPublished - 2011
EventUCNLG+Eval: Language Generation and Evaluation Workshop - Edinburgh, United Kingdom
Duration: 21 Jul 2011 → …

Conference

ConferenceUCNLG+Eval: Language Generation and Evaluation Workshop
CountryUnited Kingdom
CityEdinburgh
Period21/07/11 → …

    Fingerprint

Cite this

Reiter, E. B. (2011). Task-Based Evaluation of NLG Systems: Control vs Real-World Context. In Proceedings of UCNLG+Eval 2011 (pp. 28-32). Association for Computational Linguistics. http://aclweb.org/anthology-new/W/W11/W11-2704.pdf