INSPIRED: Toward sociable recommendation dialog systems

Shirley Anugrah Hayati; Dongyeop Kang; Qingxiaoyang Zhu; Weiyan Shi; Zhou Yu

INSPIRED: Toward sociable recommendation dialog systems

Shirley Anugrah Hayati
, Dongyeop Kang
, Qingxiaoyang Zhu
, Weiyan Shi
, Zhou Yu

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

Abstract

In recommendation dialogs, humans commonly disclose their preference and make recommendations in a friendly manner. However, this is a challenge in developing a sociable recommendation dialog system, due to the lack of dialog dataset annotated with such sociable strategies. Therefore, we present INSPIRED, a new dataset of 1,001 human-human dialogs for movie recommendation with measures for successful recommendations. To better understand how humans make recommendations in communication, we design an annotation scheme related to recommendation strategies based on social science theories and annotate these dialogs. Our analysis shows that sociable recommendation strategies, such as sharing personal opinions or communicating with encouragement, more frequently lead to successful recommendations. Based on our dataset, we train end-to-end recommendation dialog systems with and without our strategy labels. In both automatic and human evaluation, our model with strategy incorporation outperforms the baseline model. This work is a first step for building sociable recommendation dialog systems with a basis of social science theories.

Original language	English (US)
Title of host publication	EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
Publisher	Association for Computational Linguistics (ACL)
Pages	8142-8152
Number of pages	11
ISBN (Electronic)	9781952148606
State	Published - 2020
Externally published	Yes
Event	2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020 - Virtual, Online Duration: Nov 16 2020 → Nov 20 2020

Publication series

Name	EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference

Conference

Conference	2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020
City	Virtual, Online
Period	11/16/20 → 11/20/20

Bibliographical note

Funding Information:
We would like to thank members of the NLP lab at UC Davis for discussion and participation in the pilot study. We are also grateful for human evaluation participants and Mechanical Turk workers on contributions of building this dataset.

Publisher Copyright:
© 2020 Association for Computational Linguistics.

Find It

Find It at U of M Libraries

Cite this

Hayati, S. A., Kang, D., Zhu, Q., Shi, W., & Yu, Z. (2020). INSPIRED: Toward sociable recommendation dialog systems. In EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference (pp. 8142-8152). (EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference). Association for Computational Linguistics (ACL).

INSPIRED: Toward sociable recommendation dialog systems. / Hayati, Shirley Anugrah; Kang, Dongyeop; Zhu, Qingxiaoyang et al.
EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. Association for Computational Linguistics (ACL), 2020. p. 8142-8152 (EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference).

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

Hayati, SA, Kang, D, Zhu, Q, Shi, W & Yu, Z 2020, INSPIRED: Toward sociable recommendation dialog systems. in EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference, Association for Computational Linguistics (ACL), pp. 8142-8152, 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020, Virtual, Online, 11/16/20.

Hayati SA, Kang D, Zhu Q, Shi W, Yu Z. INSPIRED: Toward sociable recommendation dialog systems. In EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. Association for Computational Linguistics (ACL). 2020. p. 8142-8152. (EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference).

Hayati, Shirley Anugrah ; Kang, Dongyeop ; Zhu, Qingxiaoyang et al. / INSPIRED : Toward sociable recommendation dialog systems. EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. Association for Computational Linguistics (ACL), 2020. pp. 8142-8152 (EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference).

@inproceedings{8985366a8478432685eef75bc22bbb73,

title = "INSPIRED: Toward sociable recommendation dialog systems",

abstract = "In recommendation dialogs, humans commonly disclose their preference and make recommendations in a friendly manner. However, this is a challenge in developing a sociable recommendation dialog system, due to the lack of dialog dataset annotated with such sociable strategies. Therefore, we present INSPIRED, a new dataset of 1,001 human-human dialogs for movie recommendation with measures for successful recommendations. To better understand how humans make recommendations in communication, we design an annotation scheme related to recommendation strategies based on social science theories and annotate these dialogs. Our analysis shows that sociable recommendation strategies, such as sharing personal opinions or communicating with encouragement, more frequently lead to successful recommendations. Based on our dataset, we train end-to-end recommendation dialog systems with and without our strategy labels. In both automatic and human evaluation, our model with strategy incorporation outperforms the baseline model. This work is a first step for building sociable recommendation dialog systems with a basis of social science theories.",

author = "Hayati, \{Shirley Anugrah\} and Dongyeop Kang and Qingxiaoyang Zhu and Weiyan Shi and Zhou Yu",

note = "Funding Information: We would like to thank members of the NLP lab at UC Davis for discussion and participation in the pilot study. We are also grateful for human evaluation participants and Mechanical Turk workers on contributions of building this dataset. Publisher Copyright: {\textcopyright} 2020 Association for Computational Linguistics.; 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020 ; Conference date: 16-11-2020 Through 20-11-2020",

year = "2020",

language = "English (US)",

series = "EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference",

publisher = "Association for Computational Linguistics (ACL)",

pages = "8142--8152",

booktitle = "EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference",

}

TY - GEN

T1 - INSPIRED

T2 - 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020

AU - Hayati, Shirley Anugrah

AU - Kang, Dongyeop

AU - Zhu, Qingxiaoyang

AU - Shi, Weiyan

AU - Yu, Zhou

N1 - Funding Information: We would like to thank members of the NLP lab at UC Davis for discussion and participation in the pilot study. We are also grateful for human evaluation participants and Mechanical Turk workers on contributions of building this dataset. Publisher Copyright: © 2020 Association for Computational Linguistics.

PY - 2020

Y1 - 2020

N2 - In recommendation dialogs, humans commonly disclose their preference and make recommendations in a friendly manner. However, this is a challenge in developing a sociable recommendation dialog system, due to the lack of dialog dataset annotated with such sociable strategies. Therefore, we present INSPIRED, a new dataset of 1,001 human-human dialogs for movie recommendation with measures for successful recommendations. To better understand how humans make recommendations in communication, we design an annotation scheme related to recommendation strategies based on social science theories and annotate these dialogs. Our analysis shows that sociable recommendation strategies, such as sharing personal opinions or communicating with encouragement, more frequently lead to successful recommendations. Based on our dataset, we train end-to-end recommendation dialog systems with and without our strategy labels. In both automatic and human evaluation, our model with strategy incorporation outperforms the baseline model. This work is a first step for building sociable recommendation dialog systems with a basis of social science theories.

AB - In recommendation dialogs, humans commonly disclose their preference and make recommendations in a friendly manner. However, this is a challenge in developing a sociable recommendation dialog system, due to the lack of dialog dataset annotated with such sociable strategies. Therefore, we present INSPIRED, a new dataset of 1,001 human-human dialogs for movie recommendation with measures for successful recommendations. To better understand how humans make recommendations in communication, we design an annotation scheme related to recommendation strategies based on social science theories and annotate these dialogs. Our analysis shows that sociable recommendation strategies, such as sharing personal opinions or communicating with encouragement, more frequently lead to successful recommendations. Based on our dataset, we train end-to-end recommendation dialog systems with and without our strategy labels. In both automatic and human evaluation, our model with strategy incorporation outperforms the baseline model. This work is a first step for building sociable recommendation dialog systems with a basis of social science theories.

UR - https://www.scopus.com/pages/publications/85101036272

UR - https://www.scopus.com/pages/publications/85101036272#tab=citedBy

M3 - Conference contribution

AN - SCOPUS:85101036272

T3 - EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference

SP - 8142

EP - 8152

BT - EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference

PB - Association for Computational Linguistics (ACL)

Y2 - 16 November 2020 through 20 November 2020

ER -

INSPIRED: Toward sociable recommendation dialog systems

Abstract

Publication series

Conference

Bibliographical note

Find It

Other files and links

Fingerprint

Cite this