Detecting ideological bias in news Websites

Autor: Aires, Victoria Patricia Silva
Přispěvatelé: Nakamura, Fabiola Guerra, Silva, Altigran Soares da, Freire, Juliana
Jazyk: portugalština
Rok vydání: 2020
Předmět:
Zdroj: Biblioteca Digital de Teses e Dissertações da UFAM
Universidade Federal do Amazonas (UFAM)
instacron:UFAM
Popis: Submitted by Victoria Aires (victoria.aires@icomp.ufam.edu.br) on 2020-09-11T16:15:08Z No. of bitstreams: 4 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) dissertacaoVictoria-vfinal-PosCorrecoes.pdf: 2577589 bytes, checksum: 315659f8a447ca2ecf620cd1c084fa2a (MD5) 359 folha de aprovac??a??o - victoria aires.pdf: 1013230 bytes, checksum: 02f7b35ce226bb4b4fdfa643215ac259 (MD5) cartaencaminhamentoautodeposito.pdf: 117236 bytes, checksum: a42dfb8baae7768f7b98b58a078c0214 (MD5) Rejected by PPGI Inform??tica (secretariappgi@icomp.ufam.edu.br), reason: Boa tarde, Na vers??o final ?? necess??rio gerar a ficha catalogr??fica e inserir entre a Folha de Rosto e a Folha de Aprova????o. on 2020-09-14T16:17:45Z (GMT) Submitted by Victoria Aires (victoria.aires@icomp.ufam.edu.br) on 2020-09-14T23:21:54Z No. of bitstreams: 4 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) 359 folha de aprovac??a??o - victoria aires.pdf: 1013230 bytes, checksum: 02f7b35ce226bb4b4fdfa643215ac259 (MD5) cartaencaminhamentoautodeposito.pdf: 117236 bytes, checksum: a42dfb8baae7768f7b98b58a078c0214 (MD5) dissertacaoVictoria-vfinal-PosCorrecoes-folha.pdf: 2870200 bytes, checksum: 96367b78bf44566c09541cf7687fe790 (MD5) Approved for entry into archive by PPGI Inform??tica (secretariappgi@icomp.ufam.edu.br) on 2020-09-15T18:26:15Z (GMT) No. of bitstreams: 4 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) 359 folha de aprovac??a??o - victoria aires.pdf: 1013230 bytes, checksum: 02f7b35ce226bb4b4fdfa643215ac259 (MD5) cartaencaminhamentoautodeposito.pdf: 117236 bytes, checksum: a42dfb8baae7768f7b98b58a078c0214 (MD5) dissertacaoVictoria-vfinal-PosCorrecoes-folha.pdf: 2870200 bytes, checksum: 96367b78bf44566c09541cf7687fe790 (MD5) Approved for entry into archive by Divis??o de Documenta????o/BC Biblioteca Central (ddbc@ufam.edu.br) on 2020-09-16T13:30:22Z (GMT) No. of bitstreams: 4 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) 359 folha de aprovac??a??o - victoria aires.pdf: 1013230 bytes, checksum: 02f7b35ce226bb4b4fdfa643215ac259 (MD5) cartaencaminhamentoautodeposito.pdf: 117236 bytes, checksum: a42dfb8baae7768f7b98b58a078c0214 (MD5) dissertacaoVictoria-vfinal-PosCorrecoes-folha.pdf: 2870200 bytes, checksum: 96367b78bf44566c09541cf7687fe790 (MD5) Made available in DSpace on 2020-09-16T13:30:22Z (GMT). No. of bitstreams: 4 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) 359 folha de aprovac??a??o - victoria aires.pdf: 1013230 bytes, checksum: 02f7b35ce226bb4b4fdfa643215ac259 (MD5) cartaencaminhamentoautodeposito.pdf: 117236 bytes, checksum: a42dfb8baae7768f7b98b58a078c0214 (MD5) dissertacaoVictoria-vfinal-PosCorrecoes-folha.pdf: 2870200 bytes, checksum: 96367b78bf44566c09541cf7687fe790 (MD5) Previous issue date: 2020-08-10 CAPES - Coordena????o de Aperfei??oamento de Pessoal de N??vel Superior Nowadays, websites or news portals are the main sources of information to most people. However, like traditional media, these vehicles can have a bias in the way they report news, favoring an ideology of interest. Combined with social media and the ease of spreading this type of content, this fact strongly contributes to polarization, hate crimes and other consequences in public opinion. To make the information more transparent to the public, it is necessary to develop methods to characterize the ideological orientation/leaning of these portals automatically. Recent approaches are not exactly suitable for this problem, as they mostly depend on external sources, generating inaccurate results otherwise. Therefore, in this work, we present methods to detect ideological/political bias in news portals based only on news articles from these portals, without any external sources. We developed two approaches: exploring hyperlinks and textual content. The objective is to demonstrate the efficiency and effectiveness of this strategy compared to the current literature. As a result, we show that an approach based on hyperlinks is capable of detecting ideological biases in a polarized scenario through a method based on citation patterns. In addition, we present an approach based on textual content associated with Information Theory concepts and show that the method is able to overcome a more traditional baseline, obtaining almost twice the accuracy/F1 in three datasets and three distinct classification tasks (bi-class and multi-class), while employing a set of only four features (against 282 employed by the baseline) when detecting different levels of ideological bias in news portals. Nos dias atuais, websites ou portais de not??cias s??o os principais meios pelos quais as pessoas consomem informa????o. Entretanto, assim como m??dias tradicionais, esses ve??culos podem ter um vi??s na maneira como reportam not??cias, favorecendo uma ideologia de interesse. Combinado ??s m??dias sociais e ?? facilidade de compartilhamento e alcance desse tipo de conte??do, esse fato contribui fortemente para a polariza????o, crimes de ??dio e outras consequ??ncias na opini??o p??blica. Para tornar as informa????es mais transparentes ao p??blico, ?? necess??rio desenvolver m??todos para caracterizar a orienta????o ideol??gica destes portais automaticamente. Abordagens propostas recentemente n??o s??o exatamente adequadas para este problema, pois dependem, em sua maioria, de fontes externas, gerando resultados imprecisos caso contr??rio. Diante disso, neste trabalho apresentamos m??todos para detectar vi??s ideol??gico em portais de not??cias baseado apenas nos artigos de not??cias oriundos destes portais, sem nenhuma fonte externa. Exploramos duas abordagens: an??lise de hiperlinks e conte??do textual. O objetivo ?? demonstrar a efici??ncia e efic??cia dessa estrat??gia comparada ?? literatura atual. Como resultados, mostramos que uma abordagem baseada em hiperlinks ?? capaz de detectar vi??s ideol??gico em um cen??rio polarizado atrav??s de um m??todo baseada em padr??es de cita????es. Al??m disso, apresentamos uma abordagem baseada em conte??do textual associada a conceitos de Teoria da Informa????o e mostramos que o m??todo ?? capaz de superar um baseline mais tradicional, obtendo quase o dobro de acur??cia/F1 em tr??s bases de dados e tr??s tarefas de classifica????o diferentes (bi-classe e multi-classe), enquanto emprega um conjunto de apenas quatro atributos (contra 282 do baseline) na detec????o de diferentes n??veis de vi??s ideol??gico em portais de not??cias.
Databáze: OpenAIRE
Externí odkaz: