Choosing a set of coherence relations for text generation: A data-driven approach

Alistair Knott; Robert Dale

Choosing a set of coherence relations for text generation: A data-driven approach

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

Abstract

An active research programme in Natural Language Generation has grown up around the notion of ‘coherence relations’. Relations are being used in a variety of roles in generation systems, in planning the structure of messages and in guiding their linguistic realisation. However, a confusing diversity exists between systems with respect to the particular relations they use. This paper outlines a systematic methodology for justifying a set of relations, on the basis of the cue phrases that can be used to signal them explicitly in text. A large taxonomy of cue phrases has been gathered, from which a conception of relations as feature-based constructs is motivated. By considering small extracts from the taxonomy, individual features of relations can then be investigated in isolation. The paper concludes by noting some advantages of a feature-theoretic treatment of relations for NLG, as well as some problems with a reliance on cue phrases in a theory of text coherence.

Original language	English
Title of host publication	Trends in Natural Language Generation: An Artificial Intelligence Perspective - Fourth European Workshop, EWNLG 1993, Selected Papers
Place of Publication	Berlin; Heidelberg
Publisher	Springer, Springer Nature
Pages	47-67
Number of pages	21
Volume	1036
ISBN (Print)	3540608001, 3540608001, 9783540608004, 9783540608004
Publication status	Published - 1996
Externally published	Yes
Event	4th European Workshop on Natural Language Generation, EWNLG 1993 - Pisa, Italy Duration: 28 Apr 1993 → 30 Apr 1993

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	1036
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Other

Other	4th European Workshop on Natural Language Generation, EWNLG 1993
Country	Italy
City	Pisa
Period	28/04/93 → 30/04/93

Access to Document

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Choosing a set of coherence relations for text generation: A data-driven approach'. Together they form a unique fingerprint.

Cite this

Knott, A., & Dale, R. (1996). Choosing a set of coherence relations for text generation: A data-driven approach. In Trends in Natural Language Generation: An Artificial Intelligence Perspective - Fourth European Workshop, EWNLG 1993, Selected Papers (Vol. 1036, pp. 47-67). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1036). Springer, Springer Nature.

Knott, Alistair ; Dale, Robert. / Choosing a set of coherence relations for text generation : A data-driven approach. Trends in Natural Language Generation: An Artificial Intelligence Perspective - Fourth European Workshop, EWNLG 1993, Selected Papers. Vol. 1036 Berlin; Heidelberg : Springer, Springer Nature, 1996. pp. 47-67 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{7c4586f94c6d4cdfb1b5b25b6400ca5d,

title = "Choosing a set of coherence relations for text generation: A data-driven approach",

abstract = "An active research programme in Natural Language Generation has grown up around the notion of {\textquoteleft}coherence relations{\textquoteright}. Relations are being used in a variety of roles in generation systems, in planning the structure of messages and in guiding their linguistic realisation. However, a confusing diversity exists between systems with respect to the particular relations they use. This paper outlines a systematic methodology for justifying a set of relations, on the basis of the cue phrases that can be used to signal them explicitly in text. A large taxonomy of cue phrases has been gathered, from which a conception of relations as feature-based constructs is motivated. By considering small extracts from the taxonomy, individual features of relations can then be investigated in isolation. The paper concludes by noting some advantages of a feature-theoretic treatment of relations for NLG, as well as some problems with a reliance on cue phrases in a theory of text coherence.",

author = "Alistair Knott and Robert Dale",

year = "1996",

language = "English",

isbn = "3540608001",

volume = "1036",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer, Springer Nature",

pages = "47--67",

booktitle = "Trends in Natural Language Generation: An Artificial Intelligence Perspective - Fourth European Workshop, EWNLG 1993, Selected Papers",

address = "United States",

note = "4th European Workshop on Natural Language Generation, EWNLG 1993 ; Conference date: 28-04-1993 Through 30-04-1993",

}

Knott, A & Dale, R 1996, Choosing a set of coherence relations for text generation: A data-driven approach. in Trends in Natural Language Generation: An Artificial Intelligence Perspective - Fourth European Workshop, EWNLG 1993, Selected Papers. vol. 1036, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1036, Springer, Springer Nature, Berlin; Heidelberg, pp. 47-67, 4th European Workshop on Natural Language Generation, EWNLG 1993, Pisa, Italy, 28/04/93.

Choosing a set of coherence relations for text generation : A data-driven approach. / Knott, Alistair; Dale, Robert.

Trends in Natural Language Generation: An Artificial Intelligence Perspective - Fourth European Workshop, EWNLG 1993, Selected Papers. Vol. 1036 Berlin; Heidelberg : Springer, Springer Nature, 1996. p. 47-67 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1036).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - Choosing a set of coherence relations for text generation

T2 - 4th European Workshop on Natural Language Generation, EWNLG 1993

AU - Knott, Alistair

AU - Dale, Robert

PY - 1996

Y1 - 1996

N2 - An active research programme in Natural Language Generation has grown up around the notion of ‘coherence relations’. Relations are being used in a variety of roles in generation systems, in planning the structure of messages and in guiding their linguistic realisation. However, a confusing diversity exists between systems with respect to the particular relations they use. This paper outlines a systematic methodology for justifying a set of relations, on the basis of the cue phrases that can be used to signal them explicitly in text. A large taxonomy of cue phrases has been gathered, from which a conception of relations as feature-based constructs is motivated. By considering small extracts from the taxonomy, individual features of relations can then be investigated in isolation. The paper concludes by noting some advantages of a feature-theoretic treatment of relations for NLG, as well as some problems with a reliance on cue phrases in a theory of text coherence.

AB - An active research programme in Natural Language Generation has grown up around the notion of ‘coherence relations’. Relations are being used in a variety of roles in generation systems, in planning the structure of messages and in guiding their linguistic realisation. However, a confusing diversity exists between systems with respect to the particular relations they use. This paper outlines a systematic methodology for justifying a set of relations, on the basis of the cue phrases that can be used to signal them explicitly in text. A large taxonomy of cue phrases has been gathered, from which a conception of relations as feature-based constructs is motivated. By considering small extracts from the taxonomy, individual features of relations can then be investigated in isolation. The paper concludes by noting some advantages of a feature-theoretic treatment of relations for NLG, as well as some problems with a reliance on cue phrases in a theory of text coherence.

UR - http://www.scopus.com/inward/record.url?scp=84876149199&partnerID=8YFLogxK

M3 - Conference proceeding contribution

AN - SCOPUS:84876149199

SN - 3540608001

SN - 9783540608004

VL - 1036

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 47

EP - 67

BT - Trends in Natural Language Generation: An Artificial Intelligence Perspective - Fourth European Workshop, EWNLG 1993, Selected Papers

PB - Springer, Springer Nature

CY - Berlin; Heidelberg

Y2 - 28 April 1993 through 30 April 1993

ER -

Knott A, Dale R. Choosing a set of coherence relations for text generation: A data-driven approach. In Trends in Natural Language Generation: An Artificial Intelligence Perspective - Fourth European Workshop, EWNLG 1993, Selected Papers. Vol. 1036. Berlin; Heidelberg: Springer, Springer Nature. 1996. p. 47-67. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).