University of Sussex
Browse

File(s) not publicly available

From RAGS to RICHES: exploiting the potential of a flexible generation architecture

presentation
posted on 2023-06-08, 09:19 authored by Lynne CahillLynne Cahill, John Carroll, Roger Evans, Daniel Paiva, Richard Power, Donia Scott, Kees van Deemter
The RAGS proposals for generic specification of NLG systems includes a detailed account of data representation, but only an outline view of processing aspects. In this paper we introduce a modular processing architecture with a concrete implementation which aims to meet the RAGS goals of transparency and reusability. We illustrate the model with the RICHES system ¿ a generation system built from simple linguisticallymotivated modules.

History

Publication status

  • Published

Publisher

Morgan Kaufmann Publishers Inc

Pages

8.0

Presentation Type

  • paper

Event name

39th Annual Meeting of the Association for Computational Linguistics

Event location

Toulouse, France

Event type

conference

ISBN

1-55860-789-7

Department affiliated with

  • Informatics Publications

Notes

Originality: Describes an instantiation of a novel, generic architecture for natural language generation systems. Rigour: Architecture realised in a substantial end-to-end generation system in the domain of patient information leaflets for drugs. Significance: Makes explicit for the first time the various types of control and data flow that occur in generation systems; exemplifies types of reuse of processing modules. Impact: The paper and related research has changed the previous, incorrect perception that generation systems are standardly organised as pipelines. Outlet: Presented at the major annual international conference on natural language processing, with acceptance rate around 20%.

Full text available

  • No

Peer reviewed?

  • Yes

Legacy Posted Date

2012-02-06

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC