Emotion recognition via random forest and galvanic skin response: Comparison of time based feature sets, window sizes and wavelet approaches

Deger Ayata; Yusuf Yaslan; Mustafa Kamasak

doi:10.1109/TIPTEKNO.2016.7863130

Emotion recognition via random forest and galvanic skin response: Comparison of time based feature sets, window sizes and wavelet approaches

Deger Ayata^*, Yusuf Yaslan, Mustafa Kamasak

^*Corresponding author for this work

Department of Computer Engineering

Istanbul Technical University

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

37 Citations (Scopus)

Abstract

Emotions play a significant and powerful role in everyday life of human beings. Developing algorithms for computers to recognize emotional expression is a widely studied area. In this study, emotion recognition from Galvanic signals was performed using time domain and wavelet based features. Feature extraction has been done with various feature set attributes. Various length windows have been used for feature extraction. Various feature attribute sets have been implemented. Valence and arousal have been categorized and relationship between physiological signals and arousal and valence has been studied using Random Forest machine learning algorithm. We have achieved 71.53% and 71.04% accuracy rate for arousal and valence respectively by using only galvanic skin response signal. We have also showed that using convolution has positive affect on accuracy rate compared to non-overlapping window based feature extraction.

Original language	English
Title of host publication	2016 Medical Technologies National Conference, TIPTEKNO 2016
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781509023868
DOIs	https://doi.org/10.1109/TIPTEKNO.2016.7863130
Publication status	Published - 23 Feb 2017
Event	2016 Medical Technologies National Conference, TIPTEKNO 2016 - Antalya, Turkey Duration: 27 Oct 2016 → 29 Oct 2016

Publication series

Name	2016 Medical Technologies National Conference, TIPTEKNO 2016

Conference

Conference	2016 Medical Technologies National Conference, TIPTEKNO 2016
Country/Territory	Turkey
City	Antalya
Period	27/10/16 → 29/10/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Keywords

Biomedical Signal Processing
Emotion Recognition
Galvanic Skin Response
Machine Learning
Pattern Recognition
Physiological Signal
Random Forest

Access to Document

10.1109/TIPTEKNO.2016.7863130

Cite this

Ayata, D., Yaslan, Y., & Kamasak, M. (2017). Emotion recognition via random forest and galvanic skin response: Comparison of time based feature sets, window sizes and wavelet approaches. In 2016 Medical Technologies National Conference, TIPTEKNO 2016 Article 7863130 (2016 Medical Technologies National Conference, TIPTEKNO 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/TIPTEKNO.2016.7863130

Ayata, Deger ; Yaslan, Yusuf ; Kamasak, Mustafa. / Emotion recognition via random forest and galvanic skin response : Comparison of time based feature sets, window sizes and wavelet approaches. 2016 Medical Technologies National Conference, TIPTEKNO 2016. Institute of Electrical and Electronics Engineers Inc., 2017. (2016 Medical Technologies National Conference, TIPTEKNO 2016).

@inproceedings{00e77da835e84700b758b197a5ff0d1c,

title = "Emotion recognition via random forest and galvanic skin response: Comparison of time based feature sets, window sizes and wavelet approaches",

abstract = "Emotions play a significant and powerful role in everyday life of human beings. Developing algorithms for computers to recognize emotional expression is a widely studied area. In this study, emotion recognition from Galvanic signals was performed using time domain and wavelet based features. Feature extraction has been done with various feature set attributes. Various length windows have been used for feature extraction. Various feature attribute sets have been implemented. Valence and arousal have been categorized and relationship between physiological signals and arousal and valence has been studied using Random Forest machine learning algorithm. We have achieved 71.53% and 71.04% accuracy rate for arousal and valence respectively by using only galvanic skin response signal. We have also showed that using convolution has positive affect on accuracy rate compared to non-overlapping window based feature extraction.",

keywords = "Biomedical Signal Processing, Emotion Recognition, Galvanic Skin Response, Machine Learning, Pattern Recognition, Physiological Signal, Random Forest",

author = "Deger Ayata and Yusuf Yaslan and Mustafa Kamasak",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 2016 Medical Technologies National Conference, TIPTEKNO 2016 ; Conference date: 27-10-2016 Through 29-10-2016",

year = "2017",

month = feb,

day = "23",

doi = "10.1109/TIPTEKNO.2016.7863130",

language = "English",

series = "2016 Medical Technologies National Conference, TIPTEKNO 2016",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2016 Medical Technologies National Conference, TIPTEKNO 2016",

address = "United States",

}

Ayata, D, Yaslan, Y & Kamasak, M 2017, Emotion recognition via random forest and galvanic skin response: Comparison of time based feature sets, window sizes and wavelet approaches. in 2016 Medical Technologies National Conference, TIPTEKNO 2016., 7863130, 2016 Medical Technologies National Conference, TIPTEKNO 2016, Institute of Electrical and Electronics Engineers Inc., 2016 Medical Technologies National Conference, TIPTEKNO 2016, Antalya, Turkey, 27/10/16. https://doi.org/10.1109/TIPTEKNO.2016.7863130

Emotion recognition via random forest and galvanic skin response: Comparison of time based feature sets, window sizes and wavelet approaches. / Ayata, Deger; Yaslan, Yusuf ; Kamasak, Mustafa.
2016 Medical Technologies National Conference, TIPTEKNO 2016. Institute of Electrical and Electronics Engineers Inc., 2017. 7863130 (2016 Medical Technologies National Conference, TIPTEKNO 2016).

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

TY - GEN

T1 - Emotion recognition via random forest and galvanic skin response

T2 - 2016 Medical Technologies National Conference, TIPTEKNO 2016

AU - Ayata, Deger

AU - Yaslan, Yusuf

AU - Kamasak, Mustafa

PY - 2017/2/23

Y1 - 2017/2/23

N2 - Emotions play a significant and powerful role in everyday life of human beings. Developing algorithms for computers to recognize emotional expression is a widely studied area. In this study, emotion recognition from Galvanic signals was performed using time domain and wavelet based features. Feature extraction has been done with various feature set attributes. Various length windows have been used for feature extraction. Various feature attribute sets have been implemented. Valence and arousal have been categorized and relationship between physiological signals and arousal and valence has been studied using Random Forest machine learning algorithm. We have achieved 71.53% and 71.04% accuracy rate for arousal and valence respectively by using only galvanic skin response signal. We have also showed that using convolution has positive affect on accuracy rate compared to non-overlapping window based feature extraction.

AB - Emotions play a significant and powerful role in everyday life of human beings. Developing algorithms for computers to recognize emotional expression is a widely studied area. In this study, emotion recognition from Galvanic signals was performed using time domain and wavelet based features. Feature extraction has been done with various feature set attributes. Various length windows have been used for feature extraction. Various feature attribute sets have been implemented. Valence and arousal have been categorized and relationship between physiological signals and arousal and valence has been studied using Random Forest machine learning algorithm. We have achieved 71.53% and 71.04% accuracy rate for arousal and valence respectively by using only galvanic skin response signal. We have also showed that using convolution has positive affect on accuracy rate compared to non-overlapping window based feature extraction.

KW - Biomedical Signal Processing

KW - Emotion Recognition

KW - Galvanic Skin Response

KW - Machine Learning

KW - Pattern Recognition

KW - Physiological Signal

KW - Random Forest

UR - http://www.scopus.com/inward/record.url?scp=85016116544&partnerID=8YFLogxK

U2 - 10.1109/TIPTEKNO.2016.7863130

DO - 10.1109/TIPTEKNO.2016.7863130

M3 - Conference contribution

AN - SCOPUS:85016116544

T3 - 2016 Medical Technologies National Conference, TIPTEKNO 2016

BT - 2016 Medical Technologies National Conference, TIPTEKNO 2016

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 27 October 2016 through 29 October 2016

ER -

Ayata D, Yaslan Y , Kamasak M. Emotion recognition via random forest and galvanic skin response: Comparison of time based feature sets, window sizes and wavelet approaches. In 2016 Medical Technologies National Conference, TIPTEKNO 2016. Institute of Electrical and Electronics Engineers Inc. 2017. 7863130. (2016 Medical Technologies National Conference, TIPTEKNO 2016). doi: 10.1109/TIPTEKNO.2016.7863130

Emotion recognition via random forest and galvanic skin response: Comparison of time based feature sets, window sizes and wavelet approaches

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this