Robust clustering of data collected via crowdsourcing

Alba Pages-Zamora; Georgios B. Giannakis; Roberto Lopez-Valcarce; Pere Gimenez-Febrer

doi:10.1109/ICASSP.2017.7952910

Robust clustering of data collected via crowdsourcing

Alba Pages-Zamora, Georgios B. Giannakis, Roberto Lopez-Valcarce, Pere Gimenez-Febrer

Electrical and Computer Engineering

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

1 Scopus citations

Abstract

Crowdsourcing approaches rely on the collection of multiple individuals to solve problems that require analysis of large data sets in a timely accurate manner. The inexperience of participants or annotators motivates well robust techniques. Focusing on clustering setups, the data provided by all annotators is suitably modeled here as a mixture of Gaussian components plus a uniformly distributed random variable to capture outliers. The proposed algorithm is based on the expectation-maximization algorithm and allows for soft assignments of data to clusters, to rate annotators according to their performance, and to estimate the number of Gaussian components in the non-Gaussian/Gaussian mixture model, in a jointly manner.

Original language	English (US)
Title of host publication	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	4014-4018
Number of pages	5
ISBN (Electronic)	9781509041176
DOIs	https://doi.org/10.1109/ICASSP.2017.7952910
State	Published - Jun 16 2017
Event	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - New Orleans, United States Duration: Mar 5 2017 → Mar 9 2017

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)	1520-6149

Other

Other	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017
Country/Territory	United States
City	New Orleans
Period	3/5/17 → 3/9/17

Bibliographical note

Funding Information:
This work has been funded by the Ministerio de Economia y Competitividad of the Spanish Government, ERDF funds (TEC2013-41315-R,TEC2015-69648-REDC, TEC2016-75067-C4-2-R,TEC2013-47020-C2-1-R, TACTICA), the Catalan Government (2014 SGR 60 AGAUR), and the Galician Government (AtlantTIC, GRC2013/009, R2014/037).

Publisher Copyright:
© 2017 IEEE.

Keywords

Bayesian Information Criterion
Crowdsourcing
EM algorithm
Gaussian plus non-Gaussian Mixture
Outlier

Publisher link

10.1109/ICASSP.2017.7952910

Find It

Find It at U of M Libraries

Cite this

Pages-Zamora, A., Giannakis, G. B., Lopez-Valcarce, R., & Gimenez-Febrer, P. (2017). Robust clustering of data collected via crowdsourcing. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings (pp. 4014-4018). [7952910] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2017.7952910

Robust clustering of data collected via crowdsourcing. / Pages-Zamora, Alba; Giannakis, Georgios B.; Lopez-Valcarce, Roberto et al.

2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 4014-4018 7952910 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

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

Pages-Zamora, A, Giannakis, GB, Lopez-Valcarce, R & Gimenez-Febrer, P 2017, Robust clustering of data collected via crowdsourcing. in 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings., 7952910, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 4014-4018, 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017, New Orleans, United States, 3/5/17. https://doi.org/10.1109/ICASSP.2017.7952910

Pages-Zamora A, Giannakis GB, Lopez-Valcarce R, Gimenez-Febrer P. Robust clustering of data collected via crowdsourcing. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 4014-4018. 7952910. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2017.7952910

Pages-Zamora, Alba ; Giannakis, Georgios B. ; Lopez-Valcarce, Roberto et al. / Robust clustering of data collected via crowdsourcing. 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 4014-4018 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{1a7a3294bc30489486b3b1ddfe192a77,

title = "Robust clustering of data collected via crowdsourcing",

abstract = "Crowdsourcing approaches rely on the collection of multiple individuals to solve problems that require analysis of large data sets in a timely accurate manner. The inexperience of participants or annotators motivates well robust techniques. Focusing on clustering setups, the data provided by all annotators is suitably modeled here as a mixture of Gaussian components plus a uniformly distributed random variable to capture outliers. The proposed algorithm is based on the expectation-maximization algorithm and allows for soft assignments of data to clusters, to rate annotators according to their performance, and to estimate the number of Gaussian components in the non-Gaussian/Gaussian mixture model, in a jointly manner.",

keywords = "Bayesian Information Criterion, Crowdsourcing, EM algorithm, Gaussian plus non-Gaussian Mixture, Outlier",

author = "Alba Pages-Zamora and Giannakis, {Georgios B.} and Roberto Lopez-Valcarce and Pere Gimenez-Febrer",

note = "Funding Information: This work has been funded by the Ministerio de Economia y Competitividad of the Spanish Government, ERDF funds (TEC2013-41315-R,TEC2015-69648-REDC, TEC2016-75067-C4-2-R,TEC2013-47020-C2-1-R, TACTICA), the Catalan Government (2014 SGR 60 AGAUR), and the Galician Government (AtlantTIC, GRC2013/009, R2014/037). Publisher Copyright: {\textcopyright} 2017 IEEE.; 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 ; Conference date: 05-03-2017 Through 09-03-2017",

year = "2017",

month = jun,

day = "16",

doi = "10.1109/ICASSP.2017.7952910",

language = "English (US)",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

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

pages = "4014--4018",

booktitle = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings",

}

TY - GEN

T1 - Robust clustering of data collected via crowdsourcing

AU - Pages-Zamora, Alba

AU - Giannakis, Georgios B.

AU - Lopez-Valcarce, Roberto

AU - Gimenez-Febrer, Pere

N1 - Funding Information: This work has been funded by the Ministerio de Economia y Competitividad of the Spanish Government, ERDF funds (TEC2013-41315-R,TEC2015-69648-REDC, TEC2016-75067-C4-2-R,TEC2013-47020-C2-1-R, TACTICA), the Catalan Government (2014 SGR 60 AGAUR), and the Galician Government (AtlantTIC, GRC2013/009, R2014/037). Publisher Copyright: © 2017 IEEE.

PY - 2017/6/16

Y1 - 2017/6/16

N2 - Crowdsourcing approaches rely on the collection of multiple individuals to solve problems that require analysis of large data sets in a timely accurate manner. The inexperience of participants or annotators motivates well robust techniques. Focusing on clustering setups, the data provided by all annotators is suitably modeled here as a mixture of Gaussian components plus a uniformly distributed random variable to capture outliers. The proposed algorithm is based on the expectation-maximization algorithm and allows for soft assignments of data to clusters, to rate annotators according to their performance, and to estimate the number of Gaussian components in the non-Gaussian/Gaussian mixture model, in a jointly manner.

AB - Crowdsourcing approaches rely on the collection of multiple individuals to solve problems that require analysis of large data sets in a timely accurate manner. The inexperience of participants or annotators motivates well robust techniques. Focusing on clustering setups, the data provided by all annotators is suitably modeled here as a mixture of Gaussian components plus a uniformly distributed random variable to capture outliers. The proposed algorithm is based on the expectation-maximization algorithm and allows for soft assignments of data to clusters, to rate annotators according to their performance, and to estimate the number of Gaussian components in the non-Gaussian/Gaussian mixture model, in a jointly manner.

KW - Bayesian Information Criterion

KW - Crowdsourcing

KW - EM algorithm

KW - Gaussian plus non-Gaussian Mixture

KW - Outlier

UR - http://www.scopus.com/inward/record.url?scp=85023775549&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023775549&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2017.7952910

DO - 10.1109/ICASSP.2017.7952910

M3 - Conference contribution

AN - SCOPUS:85023775549

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 4014

EP - 4018

BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017

Y2 - 5 March 2017 through 9 March 2017

ER -

Robust clustering of data collected via crowdsourcing

Abstract

Publication series

Other

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this