A reference architecture for natural language generation systems

C. Mellish; D. Scott; Lynne Cahill; Roger Evans; D. Paiva; M. Reape

doi:10.1017/S1351324906004104

A reference architecture for natural language generation systems

C. Mellish, D. Scott, Lynne Cahill, Roger Evans, D. Paiva, M. Reape

University of Brighton

Research output: Contribution to journal › Article › peer-review

Abstract

We present the RAGS (Reference Architecture for Generation Systems) framework: a specification of an abstract Natural Language Generation (NLG) system architecture to support sharing, re-use, comparison and evaluation of NLG technologies. We argue that the evidence from a survey of actual NLG systems calls for a different emphasis in a reference proposal from that seen in similar initiatives in information extraction and multimedia interfaces. We introduce the framework itself, in particular the two-level data model that allows us to support the complex data requirements of NLG systems in a flexible and coherent fashion, and describe our efforts to validate the framework through a range of implementations.

Original language	English
Pages (from-to)	1-34
Number of pages	34
Journal	Natural Language Engineering
Volume	12
Issue number	1
DOIs	https://doi.org/10.1017/S1351324906004104
Publication status	Published - Mar 2006

Bibliographical note

Keywords

Natural language generation systems

Access to Document

10.1017/S1351324906004104Licence: Unspecified

A_Reference_Architecture.pdfFinal published version, 299 KBLicence: Unspecified

Cite this

@article{4f1e6933d847406680a8c053d10bf965,

title = "A reference architecture for natural language generation systems",

abstract = "We present the RAGS (Reference Architecture for Generation Systems) framework: a specification of an abstract Natural Language Generation (NLG) system architecture to support sharing, re-use, comparison and evaluation of NLG technologies. We argue that the evidence from a survey of actual NLG systems calls for a different emphasis in a reference proposal from that seen in similar initiatives in information extraction and multimedia interfaces. We introduce the framework itself, in particular the two-level data model that allows us to support the complex data requirements of NLG systems in a flexible and coherent fashion, and describe our efforts to validate the framework through a range of implementations.",

keywords = "Natural language generation systems",

author = "C. Mellish and D. Scott and Lynne Cahill and Roger Evans and D. Paiva and M. Reape",

note = "{\textcopyright} 2006 Cambridge University Press",

year = "2006",

month = mar,

doi = "10.1017/S1351324906004104",

language = "English",

volume = "12",

pages = "1--34",

journal = "Natural Language Engineering",

issn = "1469-8110",

number = "1",

}

TY - JOUR

T1 - A reference architecture for natural language generation systems

AU - Mellish, C.

AU - Scott, D.

AU - Cahill, Lynne

AU - Evans, Roger

AU - Paiva, D.

AU - Reape, M.

PY - 2006/3

Y1 - 2006/3

N2 - We present the RAGS (Reference Architecture for Generation Systems) framework: a specification of an abstract Natural Language Generation (NLG) system architecture to support sharing, re-use, comparison and evaluation of NLG technologies. We argue that the evidence from a survey of actual NLG systems calls for a different emphasis in a reference proposal from that seen in similar initiatives in information extraction and multimedia interfaces. We introduce the framework itself, in particular the two-level data model that allows us to support the complex data requirements of NLG systems in a flexible and coherent fashion, and describe our efforts to validate the framework through a range of implementations.

AB - We present the RAGS (Reference Architecture for Generation Systems) framework: a specification of an abstract Natural Language Generation (NLG) system architecture to support sharing, re-use, comparison and evaluation of NLG technologies. We argue that the evidence from a survey of actual NLG systems calls for a different emphasis in a reference proposal from that seen in similar initiatives in information extraction and multimedia interfaces. We introduce the framework itself, in particular the two-level data model that allows us to support the complex data requirements of NLG systems in a flexible and coherent fashion, and describe our efforts to validate the framework through a range of implementations.

KW - Natural language generation systems

U2 - 10.1017/S1351324906004104

DO - 10.1017/S1351324906004104

M3 - Article

SN - 1469-8110

VL - 12

SP - 1

EP - 34

JO - Natural Language Engineering

JF - Natural Language Engineering

IS - 1

ER -

A reference architecture for natural language generation systems

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this