A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification

Ruizhe Li; Chenghua Lin; Matthew Collinson; Xiao Li; Guanyi Chen

doi:10.18653/v1/K19-1036

A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification

Ruizhe Li^*, Chenghua Lin, Matthew Collinson, Xiao Li, Guanyi Chen

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Published conference contribution

20 Citations (Scopus)

18 Downloads (Pure)

Abstract

Recognising dialogue acts (DA) is important for many natural language processing tasks such as dialogue generation and intention recognition. In this paper, we propose a dual-attention hierarchical recurrent neural network for DA classification. Our model is partially inspired by the observation that conversational utterances are normally associated with both a DA and a topic, where the former captures the social act and the latter describes the subject matter. However, such a dependency between DAs and topics has not been utilised by most existing systems for DA classification. With a novel dual task-specific attention mechanism, our model is able, for utterances, to capture information about both DAs and topics, as well as information about the interactions between them. Experimental results show that by modelling topic as an auxiliary task, our model can significantly improve DA classification, yielding better or comparable performance to the state-of-the-art method on three public datasets.

Original language	English
Title of host publication	Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)
Editors	Mohit Bansal, Aline Villavicencio
Place of Publication	Hong Kong, China
Publisher	Association for Computational Linguistics
Pages	383-392
Number of pages	10
ISBN (Electronic)	9781950737727
ISBN (Print)	978-1-950737-72-7
DOIs	https://doi.org/10.18653/v1/K19-1036
Publication status	Published - Nov 2019
Event	23rd Conference on Computational Natural Language Learning, CoNLL 2019 - Hong Kong, China Duration: 3 Nov 2019 → 4 Nov 2019

Publication series

Name	CoNLL 2019 - 23rd Conference on Computational Natural Language Learning, Proceedings of the Conference

Conference

Conference	23rd Conference on Computational Natural Language Learning, CoNLL 2019
Country/Territory	China
City	Hong Kong
Period	3/11/19 → 4/11/19

Keywords

cs.CL

Access to Document

10.18653/v1/K19-1036Licence: CC BY

1810.09154v3Submitted manuscript, 1.09 MBLicence: Unspecified
Li_etal_Dual_Attention_VOR
https://creativecommons.org/licenses/by/4.0/
Final published version, 1.09 MBLicence: CC BY

https://arxiv.org/abs/1810.09154v3Licence: Unspecified
https://www.aclweb.org/anthology/K19-1036.pdfLicence: CC BY

Cite this

Li, R., Lin, C., Collinson, M., Li, X., & Chen, G. (2019). A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification. In M. Bansal, & A. Villavicencio (Eds.), Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL) (pp. 383-392). (CoNLL 2019 - 23rd Conference on Computational Natural Language Learning, Proceedings of the Conference). Association for Computational Linguistics. https://doi.org/10.18653/v1/K19-1036

A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification. / Li, Ruizhe; Lin, Chenghua; Collinson, Matthew et al.

Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL). ed. / Mohit Bansal; Aline Villavicencio. Hong Kong, China : Association for Computational Linguistics, 2019. p. 383-392 (CoNLL 2019 - 23rd Conference on Computational Natural Language Learning, Proceedings of the Conference).

Research output: Chapter in Book/Report/Conference proceeding › Published conference contribution

Li, R, Lin, C, Collinson, M, Li, X & Chen, G 2019, A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification. in M Bansal & A Villavicencio (eds), Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL). CoNLL 2019 - 23rd Conference on Computational Natural Language Learning, Proceedings of the Conference, Association for Computational Linguistics, Hong Kong, China, pp. 383-392, 23rd Conference on Computational Natural Language Learning, CoNLL 2019, Hong Kong, China, 3/11/19. https://doi.org/10.18653/v1/K19-1036

Li R, Lin C, Collinson M, Li X, Chen G. A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification. In Bansal M, Villavicencio A, editors, Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL). Hong Kong, China: Association for Computational Linguistics. 2019. p. 383-392. (CoNLL 2019 - 23rd Conference on Computational Natural Language Learning, Proceedings of the Conference). doi: 10.18653/v1/K19-1036

Li, Ruizhe ; Lin, Chenghua ; Collinson, Matthew et al. / A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification. Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL). editor / Mohit Bansal ; Aline Villavicencio. Hong Kong, China : Association for Computational Linguistics, 2019. pp. 383-392 (CoNLL 2019 - 23rd Conference on Computational Natural Language Learning, Proceedings of the Conference).

@inproceedings{e9e45269f3544157a4e7186a26ad808b,

title = "A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification",

abstract = "Recognising dialogue acts (DA) is important for many natural language processing tasks such as dialogue generation and intention recognition. In this paper, we propose a dual-attention hierarchical recurrent neural network for DA classification. Our model is partially inspired by the observation that conversational utterances are normally associated with both a DA and a topic, where the former captures the social act and the latter describes the subject matter. However, such a dependency between DAs and topics has not been utilised by most existing systems for DA classification. With a novel dual task-specific attention mechanism, our model is able, for utterances, to capture information about both DAs and topics, as well as information about the interactions between them. Experimental results show that by modelling topic as an auxiliary task, our model can significantly improve DA classification, yielding better or comparable performance to the state-of-the-art method on three public datasets.",

keywords = "cs.CL",

author = "Ruizhe Li and Chenghua Lin and Matthew Collinson and Xiao Li and Guanyi Chen",

note = "Acknowledgment This work is supported by the award made by the UK Engineering and Physical Sciences Research Council (Grant number: EP/P011829/1).; 23rd Conference on Computational Natural Language Learning, CoNLL 2019 ; Conference date: 03-11-2019 Through 04-11-2019",

year = "2019",

month = nov,

doi = "10.18653/v1/K19-1036",

language = "English",

isbn = "978-1-950737-72-7",

series = "CoNLL 2019 - 23rd Conference on Computational Natural Language Learning, Proceedings of the Conference",

publisher = "Association for Computational Linguistics",

pages = "383--392",

editor = "Bansal, {Mohit } and Villavicencio, {Aline }",

booktitle = "Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)",

}

TY - GEN

T1 - A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification

AU - Li, Ruizhe

AU - Lin, Chenghua

AU - Collinson, Matthew

AU - Li, Xiao

AU - Chen, Guanyi

N1 - Acknowledgment This work is supported by the award made by the UK Engineering and Physical Sciences Research Council (Grant number: EP/P011829/1).

PY - 2019/11

Y1 - 2019/11

N2 - Recognising dialogue acts (DA) is important for many natural language processing tasks such as dialogue generation and intention recognition. In this paper, we propose a dual-attention hierarchical recurrent neural network for DA classification. Our model is partially inspired by the observation that conversational utterances are normally associated with both a DA and a topic, where the former captures the social act and the latter describes the subject matter. However, such a dependency between DAs and topics has not been utilised by most existing systems for DA classification. With a novel dual task-specific attention mechanism, our model is able, for utterances, to capture information about both DAs and topics, as well as information about the interactions between them. Experimental results show that by modelling topic as an auxiliary task, our model can significantly improve DA classification, yielding better or comparable performance to the state-of-the-art method on three public datasets.

AB - Recognising dialogue acts (DA) is important for many natural language processing tasks such as dialogue generation and intention recognition. In this paper, we propose a dual-attention hierarchical recurrent neural network for DA classification. Our model is partially inspired by the observation that conversational utterances are normally associated with both a DA and a topic, where the former captures the social act and the latter describes the subject matter. However, such a dependency between DAs and topics has not been utilised by most existing systems for DA classification. With a novel dual task-specific attention mechanism, our model is able, for utterances, to capture information about both DAs and topics, as well as information about the interactions between them. Experimental results show that by modelling topic as an auxiliary task, our model can significantly improve DA classification, yielding better or comparable performance to the state-of-the-art method on three public datasets.

KW - cs.CL

UR - https://www.aclweb.org/anthology/K19-1/

U2 - 10.18653/v1/K19-1036

DO - 10.18653/v1/K19-1036

M3 - Published conference contribution

SN - 978-1-950737-72-7

T3 - CoNLL 2019 - 23rd Conference on Computational Natural Language Learning, Proceedings of the Conference

SP - 383

EP - 392

BT - Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)

A2 - Bansal, Mohit

A2 - Villavicencio, Aline

PB - Association for Computational Linguistics

CY - Hong Kong, China

T2 - 23rd Conference on Computational Natural Language Learning, CoNLL 2019

Y2 - 3 November 2019 through 4 November 2019

ER -

A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Matthew Collinson

Cite this

A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Profiles

Matthew Collinson

Cite this