Online attention for interpretable conflict estimation in political debates

Ruben Vereecken; Stavros Petridis; Yiannis Panagakis; Maja Pantic

doi:10.1109/FG.2018.00062

Online attention for interpretable conflict estimation in political debates

Ruben Vereecken
, Stavros Petridis
, Yiannis Panagakis
, Maja Pantic

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

1 Citation (Scopus)

12 Downloads (Pure)

Abstract

Conflict arises naturally in dyadic interactions when involved individuals act on incompatible goals, interests, or actions. In this paper, the problem of conflict intensity estimation from audiovisual recordings is addressed. To this end, we propose an online attention-based neural network in order to learn a mapping from a sequence of audiovisual features to time-series describing conflict intensity. The proposed method is evaluated by conducting experiments in conflict intensity estimation by employing the CONFER dataset. Experimental results indicate the superiority of the proposed model compared to the state of the art. Furthermore, we demonstrate that by incorporating sparsity in the model, the origin of conflict can be traced back to specific key frames facilitating the interpretation of conflict escalation.

Original language	English
Title of host publication	13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018
Publisher	IEEE
Pages	389-393
Number of pages	5
ISBN (Electronic)	9781538623350
DOIs	https://doi.org/10.1109/FG.2018.00062
Publication status	Published - 5 Jun 2018
Event	13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018: First Workshop on Large-scale Emotion Recognition and Analysis - Xi'an, China Duration: 15 May 2018 → 19 May 2018 Conference number: 13 https://fg2018.cse.sc.edu/

Conference

Conference	13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018
Abbreviated title	FG 2018
Country/Territory	China
City	Xi'an
Period	15/05/18 → 19/05/18
Internet address	https://fg2018.cse.sc.edu/

Keywords

Attention
CONFER
Conflict
Debate
Estimation
Network
Political

Access to Document

10.1109/FG.2018.00062

Cite this

@inproceedings{916f51c5969142b4bd8dd513e4b9e060,

title = "Online attention for interpretable conflict estimation in political debates",

abstract = "Conflict arises naturally in dyadic interactions when involved individuals act on incompatible goals, interests, or actions. In this paper, the problem of conflict intensity estimation from audiovisual recordings is addressed. To this end, we propose an online attention-based neural network in order to learn a mapping from a sequence of audiovisual features to time-series describing conflict intensity. The proposed method is evaluated by conducting experiments in conflict intensity estimation by employing the CONFER dataset. Experimental results indicate the superiority of the proposed model compared to the state of the art. Furthermore, we demonstrate that by incorporating sparsity in the model, the origin of conflict can be traced back to specific key frames facilitating the interpretation of conflict escalation.",

keywords = "Attention, CONFER, Conflict, Debate, Estimation, Network, Political",

author = "Ruben Vereecken and Stavros Petridis and Yiannis Panagakis and Maja Pantic",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018 : First Workshop on Large-scale Emotion Recognition and Analysis, FG 2018 ; Conference date: 15-05-2018 Through 19-05-2018",

year = "2018",

month = jun,

day = "5",

doi = "10.1109/FG.2018.00062",

language = "English",

pages = "389--393",

booktitle = "13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018",

publisher = "IEEE",

address = "United States",

url = "https://fg2018.cse.sc.edu/",

}

Vereecken, R, Petridis, S, Panagakis, Y & Pantic, M 2018, Online attention for interpretable conflict estimation in political debates. in 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018. IEEE, pp. 389-393, 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018, Xi'an, China, 15/05/18. https://doi.org/10.1109/FG.2018.00062

Online attention for interpretable conflict estimation in political debates. / Vereecken, Ruben; Petridis, Stavros; Panagakis, Yiannis et al.
13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018. IEEE, 2018. p. 389-393.

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

TY - GEN

T1 - Online attention for interpretable conflict estimation in political debates

AU - Vereecken, Ruben

AU - Petridis, Stavros

AU - Panagakis, Yiannis

AU - Pantic, Maja

N1 - Conference code: 13

PY - 2018/6/5

Y1 - 2018/6/5

N2 - Conflict arises naturally in dyadic interactions when involved individuals act on incompatible goals, interests, or actions. In this paper, the problem of conflict intensity estimation from audiovisual recordings is addressed. To this end, we propose an online attention-based neural network in order to learn a mapping from a sequence of audiovisual features to time-series describing conflict intensity. The proposed method is evaluated by conducting experiments in conflict intensity estimation by employing the CONFER dataset. Experimental results indicate the superiority of the proposed model compared to the state of the art. Furthermore, we demonstrate that by incorporating sparsity in the model, the origin of conflict can be traced back to specific key frames facilitating the interpretation of conflict escalation.

AB - Conflict arises naturally in dyadic interactions when involved individuals act on incompatible goals, interests, or actions. In this paper, the problem of conflict intensity estimation from audiovisual recordings is addressed. To this end, we propose an online attention-based neural network in order to learn a mapping from a sequence of audiovisual features to time-series describing conflict intensity. The proposed method is evaluated by conducting experiments in conflict intensity estimation by employing the CONFER dataset. Experimental results indicate the superiority of the proposed model compared to the state of the art. Furthermore, we demonstrate that by incorporating sparsity in the model, the origin of conflict can be traced back to specific key frames facilitating the interpretation of conflict escalation.

KW - Attention

KW - CONFER

KW - Conflict

KW - Debate

KW - Estimation

KW - Network

KW - Political

UR - https://www.scopus.com/pages/publications/85049404245

U2 - 10.1109/FG.2018.00062

DO - 10.1109/FG.2018.00062

M3 - Conference contribution

AN - SCOPUS:85049404245

SP - 389

EP - 393

BT - 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018

PB - IEEE

T2 - 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018

Y2 - 15 May 2018 through 19 May 2018

ER -

Online attention for interpretable conflict estimation in political debates

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this