A human in the loop approach to capture bias and support media scientists in news video analysis

Panagiotis Mavridis, Markus De Jong, Lora Aroyo, Alessandro Bozzon, Jesse De Vos, Johan Oomen, Antoaneta Dimitrova, Alec Badenoch

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

Abstract

Bias is inevitable and inherent in any form of communication. News often appear biased to citizens with different political orientations, and understood differently by news media scholars and the broader public. In this paper we advocate the need for accurate methods for bias identification in video news item, to enable rich analytics capabilities in order to assist humanities media scholars and social political scientists. We propose to analyze biases that are typical in video news (including framing, gender and racial biases) by means of a human-in-the-loop approach that combines text and image analysis with human computation techniques.

LanguageEnglish
Title of host publicationJoint Proceedings SAD 2018 and CrowdBias 2018
Subtitle of host publicationProceedings of the 1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper Proceedings of the 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management (SAD 2018 and CrowdBias 2018) co-located the 6th AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2018). Zürich, Switzerland, July 5, 2018
EditorsLora Aroyo, Anca Dumitrache
PublisherCEUR-WS
Pages88-92
Number of pages5
Publication statusPublished - 2018
Event1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management, SAD+CrowdBias 2018 - Zurich, Switzerland
Duration: 5 Jul 2018 → …

Publication series

NameCEUR Workshop Proceedings
PublisherCEUR Workshop Proceedings
Volume2276
ISSN (Print)1613-0073

Conference

Conference1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management, SAD+CrowdBias 2018
CountrySwitzerland
CityZurich
Period5/07/18 → …

Fingerprint

Image analysis
Communication

Keywords

  • Bias detection
  • bias in news video files
  • crowdsourcing
  • human computation
  • human in the loop
  • machine learning

Cite this

Mavridis, P., De Jong, M., Aroyo, L., Bozzon, A., De Vos, J., Oomen, J., ... Badenoch, A. (2018). A human in the loop approach to capture bias and support media scientists in news video analysis. In L. Aroyo, & A. Dumitrache (Eds.), Joint Proceedings SAD 2018 and CrowdBias 2018: Proceedings of the 1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper Proceedings of the 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management (SAD 2018 and CrowdBias 2018) co-located the 6th AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2018). Zürich, Switzerland, July 5, 2018 (pp. 88-92). (CEUR Workshop Proceedings; Vol. 2276). CEUR-WS.
Mavridis, Panagiotis ; De Jong, Markus ; Aroyo, Lora ; Bozzon, Alessandro ; De Vos, Jesse ; Oomen, Johan ; Dimitrova, Antoaneta ; Badenoch, Alec. / A human in the loop approach to capture bias and support media scientists in news video analysis. Joint Proceedings SAD 2018 and CrowdBias 2018: Proceedings of the 1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper Proceedings of the 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management (SAD 2018 and CrowdBias 2018) co-located the 6th AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2018). Zürich, Switzerland, July 5, 2018. editor / Lora Aroyo ; Anca Dumitrache. CEUR-WS, 2018. pp. 88-92 (CEUR Workshop Proceedings).
@inproceedings{4336fcf74b744164a0a57fc9ae04ff47,
title = "A human in the loop approach to capture bias and support media scientists in news video analysis",
abstract = "Bias is inevitable and inherent in any form of communication. News often appear biased to citizens with different political orientations, and understood differently by news media scholars and the broader public. In this paper we advocate the need for accurate methods for bias identification in video news item, to enable rich analytics capabilities in order to assist humanities media scholars and social political scientists. We propose to analyze biases that are typical in video news (including framing, gender and racial biases) by means of a human-in-the-loop approach that combines text and image analysis with human computation techniques.",
keywords = "Bias detection, bias in news video files, crowdsourcing, human computation, human in the loop, machine learning",
author = "Panagiotis Mavridis and {De Jong}, Markus and Lora Aroyo and Alessandro Bozzon and {De Vos}, Jesse and Johan Oomen and Antoaneta Dimitrova and Alec Badenoch",
year = "2018",
language = "English",
series = "CEUR Workshop Proceedings",
publisher = "CEUR-WS",
pages = "88--92",
editor = "Lora Aroyo and Anca Dumitrache",
booktitle = "Joint Proceedings SAD 2018 and CrowdBias 2018",

}

Mavridis, P, De Jong, M, Aroyo, L, Bozzon, A, De Vos, J, Oomen, J, Dimitrova, A & Badenoch, A 2018, A human in the loop approach to capture bias and support media scientists in news video analysis. in L Aroyo & A Dumitrache (eds), Joint Proceedings SAD 2018 and CrowdBias 2018: Proceedings of the 1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper Proceedings of the 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management (SAD 2018 and CrowdBias 2018) co-located the 6th AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2018). Zürich, Switzerland, July 5, 2018. CEUR Workshop Proceedings, vol. 2276, CEUR-WS, pp. 88-92, 1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management, SAD+CrowdBias 2018, Zurich, Switzerland, 5/07/18.

A human in the loop approach to capture bias and support media scientists in news video analysis. / Mavridis, Panagiotis; De Jong, Markus; Aroyo, Lora; Bozzon, Alessandro; De Vos, Jesse; Oomen, Johan; Dimitrova, Antoaneta; Badenoch, Alec.

Joint Proceedings SAD 2018 and CrowdBias 2018: Proceedings of the 1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper Proceedings of the 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management (SAD 2018 and CrowdBias 2018) co-located the 6th AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2018). Zürich, Switzerland, July 5, 2018. ed. / Lora Aroyo; Anca Dumitrache. CEUR-WS, 2018. p. 88-92 (CEUR Workshop Proceedings; Vol. 2276).

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - A human in the loop approach to capture bias and support media scientists in news video analysis

AU - Mavridis, Panagiotis

AU - De Jong, Markus

AU - Aroyo, Lora

AU - Bozzon, Alessandro

AU - De Vos, Jesse

AU - Oomen, Johan

AU - Dimitrova, Antoaneta

AU - Badenoch, Alec

PY - 2018

Y1 - 2018

N2 - Bias is inevitable and inherent in any form of communication. News often appear biased to citizens with different political orientations, and understood differently by news media scholars and the broader public. In this paper we advocate the need for accurate methods for bias identification in video news item, to enable rich analytics capabilities in order to assist humanities media scholars and social political scientists. We propose to analyze biases that are typical in video news (including framing, gender and racial biases) by means of a human-in-the-loop approach that combines text and image analysis with human computation techniques.

AB - Bias is inevitable and inherent in any form of communication. News often appear biased to citizens with different political orientations, and understood differently by news media scholars and the broader public. In this paper we advocate the need for accurate methods for bias identification in video news item, to enable rich analytics capabilities in order to assist humanities media scholars and social political scientists. We propose to analyze biases that are typical in video news (including framing, gender and racial biases) by means of a human-in-the-loop approach that combines text and image analysis with human computation techniques.

KW - Bias detection

KW - bias in news video files

KW - crowdsourcing

KW - human computation

KW - human in the loop

KW - machine learning

UR - http://www.scopus.com/inward/record.url?scp=85058945383&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058945383&partnerID=8YFLogxK

M3 - Conference contribution

T3 - CEUR Workshop Proceedings

SP - 88

EP - 92

BT - Joint Proceedings SAD 2018 and CrowdBias 2018

A2 - Aroyo, Lora

A2 - Dumitrache, Anca

PB - CEUR-WS

ER -

Mavridis P, De Jong M, Aroyo L, Bozzon A, De Vos J, Oomen J et al. A human in the loop approach to capture bias and support media scientists in news video analysis. In Aroyo L, Dumitrache A, editors, Joint Proceedings SAD 2018 and CrowdBias 2018: Proceedings of the 1st Workshop on Subjectivity, Ambiguity and Disagreement in Crowdsourcing, and Short Paper Proceedings of the 1st Workshop on Disentangling the Relation Between Crowdsourcing and Bias Management (SAD 2018 and CrowdBias 2018) co-located the 6th AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2018). Zürich, Switzerland, July 5, 2018. CEUR-WS. 2018. p. 88-92. (CEUR Workshop Proceedings).