Interaction Style Recognition Based on Multi-Layer Multi-View Profile Representation

Wen Li Wei, Jen Chun Lin, Chung Hsien Wu

研究成果: Article

摘要

Interaction Style (IS) refers to patterns of interaction containing highly contextual and innate information. Awareness of our IS can help us discover interpersonal conflicts and guide us how to interact with others. Recently, automatic IS recognition is becoming increasingly important in the design of a dialogue system for harmonious interaction. With the goal to select appropriate responses, four IS types proposed by Berens are selected as the basis for our study. In this study, multiple views (multi-views) of the utterances during interaction, including emotions and dialogue topics, are recognized first. Inspired by the emotion profile theory, the IS profiles are then extracted using the multi-view features to better characterize the IS of the interactional utterances. Similar to the multilayer architectures in deep neural networks, a multi-layer multi-view IS profile representation method, structured layer by layer through embedding the multi-views, is proposed to better interpret intermediate representations in the feature space of the interactional utterances based on a probabilistic fusion model. The IS is finally recognized by using the Support Vector Machine (SVM) based on the obtained IS profiles. Experimental results demonstrate that the proposed method achieved an encouraging IS recognition accuracy and outperformed the previous method.

原文English
文章編號7450642
頁(從 - 到)355-368
頁數14
期刊IEEE Transactions on Affective Computing
8
發行號3
DOIs
出版狀態Published - 2017 七月 1

指紋

Support vector machines
Multilayers
Fusion reactions
Deep neural networks

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction

引用此文

@article{6a8b45bce9f74aa48744ce750cbb8044,
title = "Interaction Style Recognition Based on Multi-Layer Multi-View Profile Representation",
abstract = "Interaction Style (IS) refers to patterns of interaction containing highly contextual and innate information. Awareness of our IS can help us discover interpersonal conflicts and guide us how to interact with others. Recently, automatic IS recognition is becoming increasingly important in the design of a dialogue system for harmonious interaction. With the goal to select appropriate responses, four IS types proposed by Berens are selected as the basis for our study. In this study, multiple views (multi-views) of the utterances during interaction, including emotions and dialogue topics, are recognized first. Inspired by the emotion profile theory, the IS profiles are then extracted using the multi-view features to better characterize the IS of the interactional utterances. Similar to the multilayer architectures in deep neural networks, a multi-layer multi-view IS profile representation method, structured layer by layer through embedding the multi-views, is proposed to better interpret intermediate representations in the feature space of the interactional utterances based on a probabilistic fusion model. The IS is finally recognized by using the Support Vector Machine (SVM) based on the obtained IS profiles. Experimental results demonstrate that the proposed method achieved an encouraging IS recognition accuracy and outperformed the previous method.",
author = "Wei, {Wen Li} and Lin, {Jen Chun} and Wu, {Chung Hsien}",
year = "2017",
month = "7",
day = "1",
doi = "10.1109/TAFFC.2016.2553024",
language = "English",
volume = "8",
pages = "355--368",
journal = "IEEE Transactions on Affective Computing",
issn = "1949-3045",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

Interaction Style Recognition Based on Multi-Layer Multi-View Profile Representation. / Wei, Wen Li; Lin, Jen Chun; Wu, Chung Hsien.

於: IEEE Transactions on Affective Computing, 卷 8, 編號 3, 7450642, 01.07.2017, p. 355-368.

研究成果: Article

TY - JOUR

T1 - Interaction Style Recognition Based on Multi-Layer Multi-View Profile Representation

AU - Wei, Wen Li

AU - Lin, Jen Chun

AU - Wu, Chung Hsien

PY - 2017/7/1

Y1 - 2017/7/1

N2 - Interaction Style (IS) refers to patterns of interaction containing highly contextual and innate information. Awareness of our IS can help us discover interpersonal conflicts and guide us how to interact with others. Recently, automatic IS recognition is becoming increasingly important in the design of a dialogue system for harmonious interaction. With the goal to select appropriate responses, four IS types proposed by Berens are selected as the basis for our study. In this study, multiple views (multi-views) of the utterances during interaction, including emotions and dialogue topics, are recognized first. Inspired by the emotion profile theory, the IS profiles are then extracted using the multi-view features to better characterize the IS of the interactional utterances. Similar to the multilayer architectures in deep neural networks, a multi-layer multi-view IS profile representation method, structured layer by layer through embedding the multi-views, is proposed to better interpret intermediate representations in the feature space of the interactional utterances based on a probabilistic fusion model. The IS is finally recognized by using the Support Vector Machine (SVM) based on the obtained IS profiles. Experimental results demonstrate that the proposed method achieved an encouraging IS recognition accuracy and outperformed the previous method.

AB - Interaction Style (IS) refers to patterns of interaction containing highly contextual and innate information. Awareness of our IS can help us discover interpersonal conflicts and guide us how to interact with others. Recently, automatic IS recognition is becoming increasingly important in the design of a dialogue system for harmonious interaction. With the goal to select appropriate responses, four IS types proposed by Berens are selected as the basis for our study. In this study, multiple views (multi-views) of the utterances during interaction, including emotions and dialogue topics, are recognized first. Inspired by the emotion profile theory, the IS profiles are then extracted using the multi-view features to better characterize the IS of the interactional utterances. Similar to the multilayer architectures in deep neural networks, a multi-layer multi-view IS profile representation method, structured layer by layer through embedding the multi-views, is proposed to better interpret intermediate representations in the feature space of the interactional utterances based on a probabilistic fusion model. The IS is finally recognized by using the Support Vector Machine (SVM) based on the obtained IS profiles. Experimental results demonstrate that the proposed method achieved an encouraging IS recognition accuracy and outperformed the previous method.

UR - http://www.scopus.com/inward/record.url?scp=85029940379&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029940379&partnerID=8YFLogxK

U2 - 10.1109/TAFFC.2016.2553024

DO - 10.1109/TAFFC.2016.2553024

M3 - Article

AN - SCOPUS:85029940379

VL - 8

SP - 355

EP - 368

JO - IEEE Transactions on Affective Computing

JF - IEEE Transactions on Affective Computing

SN - 1949-3045

IS - 3

M1 - 7450642

ER -