Exploiting Transformer-Based Multitask Learning for the Detection of Media Bias in News Articles

SMITS, Malte, ed.. Information for a better world : shaping the global future : 17th international conference, iConference 2022, virtual event, February 28 - March 4, 2022 : Part 1. Cham: Springer Nature, 2022, pp. 225-235. ISSN 0302-9743. eISSN 1611-3349. ISBN 978-3-030-96956-1. Available under: doi: 10.1007/978-3-030-96957-8_20

Zusammenfassung

Media has a substantial impact on the public perception of events. A one-sided or polarizing perspective on any topic is usually described as media bias. One of the ways how bias in news articles can be introduced is by altering word choice. Biased word choices are not always obvious, nor do they exhibit high context-dependency. Hence, detecting bias is often difficult. We propose a Transformer-based deep learning architecture trained via Multi-Task Learning using six bias-related data sets to tackle the media bias detection problem. Our best-performing implementation achieves a macro F1 of 0.776, a performance boost of 3% compared to our baseline, outperforming existing methods. Our results indicate Multi-Task Learning as a promising alternative to improve exist- ing baseline models in identifying slanted reporting.

Fachgebiet (DDC)

020 Bibliotheks- und Informationswissenschaft

Schlagwörter

Media bias, Text analysis, Multi-task learning, News analysis

Konferenz

iConference 2022 : Information for a Better World: shaping the global future, 28. Feb. 2022 - 4. März 2022, Virtual Event

Zitieren

ISO 690

SPINDE, Timo, Jan-David KRIEGER, Terry RUAS, Jelena MITROVIĆ, Franz GÖTZ-HAHN, Akiko AIZAWA, Bela GIPP, 2022. Exploiting Transformer-Based Multitask Learning for the Detection of Media Bias in News Articles. iConference 2022 : Information for a Better World: shaping the global future. Virtual Event, 28. Feb. 2022 - 4. März 2022. In: SMITS, Malte, ed.. Information for a better world : shaping the global future : 17th international conference, iConference 2022, virtual event, February 28 - March 4, 2022 : Part 1. Cham: Springer Nature, 2022, pp. 225-235. ISSN 0302-9743. eISSN 1611-3349. ISBN 978-3-030-96956-1. Available under: doi: 10.1007/978-3-030-96957-8_20

BibTex

@inproceedings{Spinde2022Explo-57324,
  year={2022},
  doi={10.1007/978-3-030-96957-8_20},
  title={Exploiting Transformer-Based Multitask Learning for the Detection of Media Bias in News Articles},
  number={13192},
  isbn={978-3-030-96956-1},
  issn={0302-9743},
  publisher={Springer Nature},
  address={Cham},
  series={Lecture Notes in Computer Science},
  booktitle={Information for a better world : shaping the global future : 17th international conference, iConference 2022, virtual event, February 28 - March 4, 2022 : Part 1},
  pages={225--235},
  editor={Smits, Malte},
  author={Spinde, Timo and Krieger, Jan-David and Ruas, Terry and Mitrović, Jelena and Götz-Hahn, Franz and Aizawa, Akiko and Gipp, Bela}
}

RDF

<rdf:RDF
    xmlns:dcterms="http://purl.org/dc/terms/"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:bibo="http://purl.org/ontology/bibo/"
    xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:void="http://rdfs.org/ns/void#"
    xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > 
  <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/57324">
    <dc:contributor>Ruas, Terry</dc:contributor>
    <dc:contributor>Gipp, Bela</dc:contributor>
    <dc:creator>Spinde, Timo</dc:creator>
    <dc:creator>Götz-Hahn, Franz</dc:creator>
    <dc:contributor>Götz-Hahn, Franz</dc:contributor>
    <dc:creator>Aizawa, Akiko</dc:creator>
    <dcterms:title>Exploiting Transformer-Based Multitask Learning for the Detection of Media Bias in News Articles</dcterms:title>
    <dc:contributor>Mitrović, Jelena</dc:contributor>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/46"/>
    <dc:contributor>Krieger, Jan-David</dc:contributor>
    <dc:contributor>Spinde, Timo</dc:contributor>
    <dc:creator>Gipp, Bela</dc:creator>
    <dc:language>eng</dc:language>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/57324"/>
    <dcterms:abstract xml:lang="eng">Media has a substantial impact on the public perception of events. A one-sided or polarizing perspective on any topic is usually described as media bias. One of the ways how bias in news articles can be introduced is by altering word choice. Biased word choices are not always obvious, nor do they exhibit high context-dependency. Hence, detecting bias is often difficult. We propose a Transformer-based deep learning architecture trained via Multi-Task Learning using six bias-related data sets to tackle the media bias detection problem. Our best-performing implementation achieves a macro F1 of 0.776, a performance boost of 3% compared to our baseline, outperforming existing methods. Our results indicate Multi-Task Learning as a promising alternative to improve exist- ing baseline models in identifying slanted reporting.</dcterms:abstract>
    <dcterms:issued>2022</dcterms:issued>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/46"/>
    <dc:creator>Krieger, Jan-David</dc:creator>
    <dc:creator>Ruas, Terry</dc:creator>
    <dc:creator>Mitrović, Jelena</dc:creator>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2022-04-20T09:31:57Z</dc:date>
    <dc:contributor>Aizawa, Akiko</dc:contributor>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2022-04-20T09:31:57Z</dcterms:available>
  </rdf:Description>
</rdf:RDF>

Universitätsbibliographie

Ja

Begutachtet

Ja