Estimation of user’s willingness to talk about the topic: Analysis of interviews between humans

Yuya Chiba; Akinori Ito

doi:10.1007/978-981-10-2585-3_34

Estimation of user’s willingness to talk about the topic: Analysis of interviews between humans

Yuya Chiba, Akinori Ito

ENG - Communications Engineering

Tohoku University

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

2 Citations (Scopus)

Abstract

This research tried to estimate the user’s willingness to talk about the topic provided by the dialog system. Dialog management based on the user’s willingness is assumed to improve the satisfaction the user gets from the dialog with the system. We collected interview dialogs between humans to analyze the features for estimation, and found that significant differences of the statistics of F₀ and power of the speech, and the degree of the facial movements by a statistical test. We conducted discrimination experiments by using multi-modal features with SVM, and obtained the best result when we used the audio-visual information. We obtained 80.4 % of discrimination ratio under leave-one-out condition and 77.1 % discrimination ratio under subject-open condition.

Original language	English
Title of host publication	Dialogues with Social Robots - Enablements, Analyses, and Evaluation
Editors	Kristiina Jokinen, Graham Wilcock
Publisher	Springer Verlag
Pages	411-419
Number of pages	9
ISBN (Print)	9789811025846
DOIs	https://doi.org/10.1007/978-981-10-2585-3_34
Publication status	Published - 2017
Event	7th International Workshop on Spoken Dialogue Systems, IWSDS 2016 - Saariselka, Finland Duration: 2016 Jan 13 → 2016 Jan 16

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	427 427 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	7th International Workshop on Spoken Dialogue Systems, IWSDS 2016
Country/Territory	Finland
City	Saariselka
Period	16/1/13 → 16/1/16

Keywords

Multi-modal information
Spoken dialog system
User’s willingness to talk

Access to Document

10.1007/978-981-10-2585-3_34

Cite this

Chiba, Y., & Ito, A. (2017). Estimation of user’s willingness to talk about the topic: Analysis of interviews between humans. In K. Jokinen, & G. Wilcock (Eds.), Dialogues with Social Robots - Enablements, Analyses, and Evaluation (pp. 411-419). (Lecture Notes in Electrical Engineering; Vol. 427 427 LNEE). Springer Verlag. https://doi.org/10.1007/978-981-10-2585-3_34

@inproceedings{7e7366a8d68a4699b0269e09a919f055,

title = "Estimation of user{\textquoteright}s willingness to talk about the topic: Analysis of interviews between humans",

abstract = "This research tried to estimate the user{\textquoteright}s willingness to talk about the topic provided by the dialog system. Dialog management based on the user{\textquoteright}s willingness is assumed to improve the satisfaction the user gets from the dialog with the system. We collected interview dialogs between humans to analyze the features for estimation, and found that significant differences of the statistics of F0 and power of the speech, and the degree of the facial movements by a statistical test. We conducted discrimination experiments by using multi-modal features with SVM, and obtained the best result when we used the audio-visual information. We obtained 80.4 % of discrimination ratio under leave-one-out condition and 77.1 % discrimination ratio under subject-open condition.",

keywords = "Multi-modal information, Spoken dialog system, User{\textquoteright}s willingness to talk",

author = "Yuya Chiba and Akinori Ito",

note = "Publisher Copyright: {\textcopyright} Springer Science+Business Media Singapore 2017.; 7th International Workshop on Spoken Dialogue Systems, IWSDS 2016 ; Conference date: 13-01-2016 Through 16-01-2016",

year = "2017",

doi = "10.1007/978-981-10-2585-3_34",

language = "English",

isbn = "9789811025846",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Verlag",

pages = "411--419",

editor = "Kristiina Jokinen and Graham Wilcock",

booktitle = "Dialogues with Social Robots - Enablements, Analyses, and Evaluation",

address = "Germany",

}

Chiba, Y & Ito, A 2017, Estimation of user’s willingness to talk about the topic: Analysis of interviews between humans. in K Jokinen & G Wilcock (eds), Dialogues with Social Robots - Enablements, Analyses, and Evaluation. Lecture Notes in Electrical Engineering, vol. 427 427 LNEE, Springer Verlag, pp. 411-419, 7th International Workshop on Spoken Dialogue Systems, IWSDS 2016, Saariselka, Finland, 16/1/13. https://doi.org/10.1007/978-981-10-2585-3_34

Estimation of user’s willingness to talk about the topic: Analysis of interviews between humans. / Chiba, Yuya; Ito, Akinori.
Dialogues with Social Robots - Enablements, Analyses, and Evaluation. ed. / Kristiina Jokinen; Graham Wilcock. Springer Verlag, 2017. p. 411-419 (Lecture Notes in Electrical Engineering; Vol. 427 427 LNEE).

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

TY - GEN

T1 - Estimation of user’s willingness to talk about the topic

T2 - 7th International Workshop on Spoken Dialogue Systems, IWSDS 2016

AU - Chiba, Yuya

AU - Ito, Akinori

N1 - Publisher Copyright: © Springer Science+Business Media Singapore 2017.

PY - 2017

Y1 - 2017

N2 - This research tried to estimate the user’s willingness to talk about the topic provided by the dialog system. Dialog management based on the user’s willingness is assumed to improve the satisfaction the user gets from the dialog with the system. We collected interview dialogs between humans to analyze the features for estimation, and found that significant differences of the statistics of F0 and power of the speech, and the degree of the facial movements by a statistical test. We conducted discrimination experiments by using multi-modal features with SVM, and obtained the best result when we used the audio-visual information. We obtained 80.4 % of discrimination ratio under leave-one-out condition and 77.1 % discrimination ratio under subject-open condition.

AB - This research tried to estimate the user’s willingness to talk about the topic provided by the dialog system. Dialog management based on the user’s willingness is assumed to improve the satisfaction the user gets from the dialog with the system. We collected interview dialogs between humans to analyze the features for estimation, and found that significant differences of the statistics of F0 and power of the speech, and the degree of the facial movements by a statistical test. We conducted discrimination experiments by using multi-modal features with SVM, and obtained the best result when we used the audio-visual information. We obtained 80.4 % of discrimination ratio under leave-one-out condition and 77.1 % discrimination ratio under subject-open condition.

KW - Multi-modal information

KW - Spoken dialog system

KW - User’s willingness to talk

UR - http://www.scopus.com/inward/record.url?scp=85009476470&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009476470&partnerID=8YFLogxK

U2 - 10.1007/978-981-10-2585-3_34

DO - 10.1007/978-981-10-2585-3_34

M3 - Conference contribution

AN - SCOPUS:85009476470

SN - 9789811025846

T3 - Lecture Notes in Electrical Engineering

SP - 411

EP - 419

BT - Dialogues with Social Robots - Enablements, Analyses, and Evaluation

A2 - Jokinen, Kristiina

A2 - Wilcock, Graham

PB - Springer Verlag

Y2 - 13 January 2016 through 16 January 2016

ER -

Estimation of user’s willingness to talk about the topic: Analysis of interviews between humans

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this