Action understanding based on a combination of one-versus-rest and one-versus-one multi-classification methods

Hui Liu, Wengming Zheng, Gaopeng Sun, Yanhua Shi, Yue Leng, Pan Lin, Ruimin Wang, Yuankui Yang, Jun Feng Gao, Haixian Wang, Keiji Iramina, Sheng Ge

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

When people observe the actions of others, they naturally try to understand the underlying intentions. This behavior is called action understanding, and it has an important influence on mental development, language comprehension, and socialization. In this study, we used functional near-infrared spectroscopy (fNIRS) to obtain brain signals related to action understanding and then classified different intentions. Aiming to overcome the drawbacks of traditional multiclass classification methods of one-versus-rest (OVR) and one-versus-one (OVO), in this paper, we propose a new effective method to solve multiclass classification that is a combination of OVR and OVO. Compared with OVO, this new method effectively improved the accuracy of four-class classification from 25% to 48%.

Original languageEnglish
Title of host publicationProceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017
EditorsSong Qiu, Hongying Liu, Li Sun, Lipo Wang, Qingli Li, Mei Zhou
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-5
Number of pages5
ISBN (Electronic)9781538619377
DOIs
Publication statusPublished - Feb 22 2018
Event10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017 - Shanghai, China
Duration: Oct 14 2017Oct 16 2017

Publication series

NameProceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017
Volume2018-January

Other

Other10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017
CountryChina
CityShanghai
Period10/14/1710/16/17

Fingerprint

Language Development
Near infrared spectroscopy
Near-Infrared Spectroscopy
Socialization
Brain

All Science Journal Classification (ASJC) codes

  • Health Informatics
  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Biomedical Engineering

Cite this

Liu, H., Zheng, W., Sun, G., Shi, Y., Leng, Y., Lin, P., ... Ge, S. (2018). Action understanding based on a combination of one-versus-rest and one-versus-one multi-classification methods. In S. Qiu, H. Liu, L. Sun, L. Wang, Q. Li, & M. Zhou (Eds.), Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017 (pp. 1-5). (Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CISP-BMEI.2017.8302159

Action understanding based on a combination of one-versus-rest and one-versus-one multi-classification methods. / Liu, Hui; Zheng, Wengming; Sun, Gaopeng; Shi, Yanhua; Leng, Yue; Lin, Pan; Wang, Ruimin; Yang, Yuankui; Gao, Jun Feng; Wang, Haixian; Iramina, Keiji; Ge, Sheng.

Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017. ed. / Song Qiu; Hongying Liu; Li Sun; Lipo Wang; Qingli Li; Mei Zhou. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1-5 (Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017; Vol. 2018-January).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Liu, H, Zheng, W, Sun, G, Shi, Y, Leng, Y, Lin, P, Wang, R, Yang, Y, Gao, JF, Wang, H, Iramina, K & Ge, S 2018, Action understanding based on a combination of one-versus-rest and one-versus-one multi-classification methods. in S Qiu, H Liu, L Sun, L Wang, Q Li & M Zhou (eds), Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017. Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 1-5, 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017, Shanghai, China, 10/14/17. https://doi.org/10.1109/CISP-BMEI.2017.8302159
Liu H, Zheng W, Sun G, Shi Y, Leng Y, Lin P et al. Action understanding based on a combination of one-versus-rest and one-versus-one multi-classification methods. In Qiu S, Liu H, Sun L, Wang L, Li Q, Zhou M, editors, Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1-5. (Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017). https://doi.org/10.1109/CISP-BMEI.2017.8302159
Liu, Hui ; Zheng, Wengming ; Sun, Gaopeng ; Shi, Yanhua ; Leng, Yue ; Lin, Pan ; Wang, Ruimin ; Yang, Yuankui ; Gao, Jun Feng ; Wang, Haixian ; Iramina, Keiji ; Ge, Sheng. / Action understanding based on a combination of one-versus-rest and one-versus-one multi-classification methods. Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017. editor / Song Qiu ; Hongying Liu ; Li Sun ; Lipo Wang ; Qingli Li ; Mei Zhou. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1-5 (Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017).
@inproceedings{b03346c706eb49bb92eb99bbe871fe02,
title = "Action understanding based on a combination of one-versus-rest and one-versus-one multi-classification methods",
abstract = "When people observe the actions of others, they naturally try to understand the underlying intentions. This behavior is called action understanding, and it has an important influence on mental development, language comprehension, and socialization. In this study, we used functional near-infrared spectroscopy (fNIRS) to obtain brain signals related to action understanding and then classified different intentions. Aiming to overcome the drawbacks of traditional multiclass classification methods of one-versus-rest (OVR) and one-versus-one (OVO), in this paper, we propose a new effective method to solve multiclass classification that is a combination of OVR and OVO. Compared with OVO, this new method effectively improved the accuracy of four-class classification from 25{\%} to 48{\%}.",
author = "Hui Liu and Wengming Zheng and Gaopeng Sun and Yanhua Shi and Yue Leng and Pan Lin and Ruimin Wang and Yuankui Yang and Gao, {Jun Feng} and Haixian Wang and Keiji Iramina and Sheng Ge",
year = "2018",
month = "2",
day = "22",
doi = "10.1109/CISP-BMEI.2017.8302159",
language = "English",
series = "Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1--5",
editor = "Song Qiu and Hongying Liu and Li Sun and Lipo Wang and Qingli Li and Mei Zhou",
booktitle = "Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017",
address = "United States",

}

TY - GEN

T1 - Action understanding based on a combination of one-versus-rest and one-versus-one multi-classification methods

AU - Liu, Hui

AU - Zheng, Wengming

AU - Sun, Gaopeng

AU - Shi, Yanhua

AU - Leng, Yue

AU - Lin, Pan

AU - Wang, Ruimin

AU - Yang, Yuankui

AU - Gao, Jun Feng

AU - Wang, Haixian

AU - Iramina, Keiji

AU - Ge, Sheng

PY - 2018/2/22

Y1 - 2018/2/22

N2 - When people observe the actions of others, they naturally try to understand the underlying intentions. This behavior is called action understanding, and it has an important influence on mental development, language comprehension, and socialization. In this study, we used functional near-infrared spectroscopy (fNIRS) to obtain brain signals related to action understanding and then classified different intentions. Aiming to overcome the drawbacks of traditional multiclass classification methods of one-versus-rest (OVR) and one-versus-one (OVO), in this paper, we propose a new effective method to solve multiclass classification that is a combination of OVR and OVO. Compared with OVO, this new method effectively improved the accuracy of four-class classification from 25% to 48%.

AB - When people observe the actions of others, they naturally try to understand the underlying intentions. This behavior is called action understanding, and it has an important influence on mental development, language comprehension, and socialization. In this study, we used functional near-infrared spectroscopy (fNIRS) to obtain brain signals related to action understanding and then classified different intentions. Aiming to overcome the drawbacks of traditional multiclass classification methods of one-versus-rest (OVR) and one-versus-one (OVO), in this paper, we propose a new effective method to solve multiclass classification that is a combination of OVR and OVO. Compared with OVO, this new method effectively improved the accuracy of four-class classification from 25% to 48%.

UR - http://www.scopus.com/inward/record.url?scp=85047530064&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047530064&partnerID=8YFLogxK

U2 - 10.1109/CISP-BMEI.2017.8302159

DO - 10.1109/CISP-BMEI.2017.8302159

M3 - Conference contribution

AN - SCOPUS:85047530064

T3 - Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017

SP - 1

EP - 5

BT - Proceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017

A2 - Qiu, Song

A2 - Liu, Hongying

A2 - Sun, Li

A2 - Wang, Lipo

A2 - Li, Qingli

A2 - Zhou, Mei

PB - Institute of Electrical and Electronics Engineers Inc.

ER -