Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described
Autor(a) principal: | |
---|---|
Data de Publicação: | 2024 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | por |
Título da fonte: | Rebeca (São Paulo) |
Texto Completo: | https://rebeca.socine.org.br/1/article/view/898 |
Resumo: | The audiovisual productions watched by each user on Netflix – a streaming platform - are based, in part, on data collection, computing, and archiving about how and what was previously consumed by each user and by others. Suggestions for new content are made by recommendation systems and operated by a set of algorithms, which are often kept in commercial secret. Netflix, on its website, proposes a “high-level description” of its recommendation system “in layman's language.” This article analyzes how this text explains the functioning of these tools, articulating it with authors who were already part of the group of platform’s developers, other critics, and specialists in algorithms. The analysis demonstrated that from the collection of little user data, especially if compared with the quantity of data usually extracted from social networking sites, it is possible to implement the recommendation system in an elaborated and customized way. The collected data behave as an inclusion pattern and constitute the raw material of a database that feeds the system, creating a complex personalized profile for each user. This profile is what recommends new titles in the search system and mainly guides the item’s position in the ranks in the initial interface. Finally, the position of the title in the interface and the row in which it is displayed significantly influence the choice of production. This, in turn, has consequences in the user’s contact with the diversity of audiovisual productions, the maintenance of the subscription, and the consumption experience on the platform. |
id |
SOCINE-1_0f8af36f06e89ad0a451aec5aefb835c |
---|---|
oai_identifier_str |
oai:ojs.emnuvens.com.br:article/898 |
network_acronym_str |
SOCINE-1 |
network_name_str |
Rebeca (São Paulo) |
repository_id_str |
|
spelling |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described Algoritmos de recomendación: un estudio sobre cómo funciona y cómo se describe el sistema de recomendaciones de Netflix Algoritmos de recomendação: um estudo sobre como funciona e como é descrito o sistema de recomendações da NetflixSistema de RecomendaçãoDadosPlataforma de StreamingAlgoritmosStreamingNetflixAlgoritmos de recomendaçãotecnologiaRecommender SystemsDataStreaming PlatformAlgorithmsRecommender SystemsDataStreaming PlatformAlgorithmsThe audiovisual productions watched by each user on Netflix – a streaming platform - are based, in part, on data collection, computing, and archiving about how and what was previously consumed by each user and by others. Suggestions for new content are made by recommendation systems and operated by a set of algorithms, which are often kept in commercial secret. Netflix, on its website, proposes a “high-level description” of its recommendation system “in layman's language.” This article analyzes how this text explains the functioning of these tools, articulating it with authors who were already part of the group of platform’s developers, other critics, and specialists in algorithms. The analysis demonstrated that from the collection of little user data, especially if compared with the quantity of data usually extracted from social networking sites, it is possible to implement the recommendation system in an elaborated and customized way. The collected data behave as an inclusion pattern and constitute the raw material of a database that feeds the system, creating a complex personalized profile for each user. This profile is what recommends new titles in the search system and mainly guides the item’s position in the ranks in the initial interface. Finally, the position of the title in the interface and the row in which it is displayed significantly influence the choice of production. This, in turn, has consequences in the user’s contact with the diversity of audiovisual productions, the maintenance of the subscription, and the consumption experience on the platform.Las producciones audiovisuales visualizadas por cada usuario en la plataforma de streaming Netflix se basan, en parte, en los datos recopilados, procesados y archivados sobre lo que fue consumido previamente por él y otros usuarios. Las sugerencias de nuevos contenidos son realizadas por sistemas de recomendación y son operacionalizadas por un conjunto de algoritmos, que a menudo se mantienen en secreto comercial. Netflix, en su sitio web, propone una "descripción de alto nivel" sobre el sistema de recomendación "en un lenguaje para personas no expertas". Este artículo analiza cómo ese texto explicita el funcionamiento de esas herramientas, relacionándolo con autores que han formado parte del grupo de programadores de la plataforma, otros críticos y expertos en algoritmos de recomendación. El análisis demostró que, a partir de la recopilación de pocos datos del usuario, especialmente en comparación con el volumen generalmente extraído de sitios de redes sociales, es posible efectuar su elaborado sistema de recomendación de forma personalizada. Los datos recopilados actúan como un "patrón de inclusión" y constituyen la materia prima de una base de datos que alimenta el sistema, creando un perfil personalizado complejo para cada individuo. Este perfil es el que recomienda nuevos títulos en el sistema de búsqueda y orienta, principalmente, la posición del elemento en las filas en la interfaz inicial. Finalmente, la posición del título en la interfaz y la fila a la que pertenece influyen significativamente en la elección de la producción, lo que tiene consecuencias en el contacto con la diversidad de productos audiovisuales, en el mantenimiento de la suscripción y en la experiencia de consumo en la plataforma.As produções audiovisuais visualizadas por cada usuário na plataforma de streaming Netflix são baseadas, em parte, nos dados coletados, tratados e arquivados sobre como e o que foi consumido anteriormente por ele e por outros usuários. As sugestões de novos conteúdos são efetuadas por sistemas de recomendação e são operacionalizadas por um conjunto de algoritmos, que por muitas vezes são mantidos em segredo comercial. A Netflix, em seu site, propõe uma “uma descrição de alto nível” sobre o sistema de recomendação “em uma linguagem para leigos”. Este artigo analisa como esse texto explicita o funcionamento dessas ferramentas, articulando-o com autores que já fizeram parte do grupo de programadores da plataforma, outros críticos, e especialistas em algoritmos de recomendação. A análise demonstrou que, a partir da coleta de poucos dados do usuário, especialmente se comprado com o volume geralmente extraído de sites de redes sociais, é possível efetivar seu elaborado sistema de recomendação de forma personalizada. Os dados coletados se comportam como um “padrão de inclusão” e se constituem em matéria prima de um banco de dados que alimenta o sistema, criando um complexo perfil personalizado para cada indivíduo. Esse perfil é o que recomenda novos títulos no sistema de busca e orienta, principalmente, a posição do item nas fileiras na interface inicial. Por fim, a posição do título na interface e a fileira da qual faz parte influenciam significativamente na escolha da produção, o que tem consequências no contato com a diversidade de produtos audiovisuais, na manutenção da assinatura, e na experiência de consumo na plataforma.Socine - Sociedade Brasileira de Estudos de Cinema e Audiovisual2024-01-13info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionArtigo avaliado pelos paresTextoinfo:eu-repo/semantics/otherapplication/pdfhttps://rebeca.socine.org.br/1/article/view/89810.22475/rebeca.v12n2.898Rebeca - Brazilian Journal for Cinema and Audiovisual Studies; Vol. 12 No. 2 (2023): Rebeca 24Rebeca - Revista Brasileira de Estudos de Cinema e Audiovisual; Vol. 12 Núm. 2 (2023): Rebeca 24Rebeca - Revista Brasileira de Estudos de Cinema e Audiovisual; Vol. 12 No 2 (2023): Rebeca 24Rebeca - Revista Brasileira de Estudos de Cinema e Audiovisual; v. 12 n. 2 (2023): Rebeca 242316-9230reponame:Rebeca (São Paulo)instname:Sociedade Brasileira de Estudos de Cinema e Audiovisual (SOCINE)instacron:SOCINEporhttps://rebeca.socine.org.br/1/article/view/898/582Copyright (c) 2023 Tiago Franklin Rodrigues Lucena, Eduarda Carretero Garcia, Mariana Maronezzi Brezovsky , Thiago Fanelli Ferraiolhttp://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccessRodrigues Lucena, Tiago FranklinGarcia, Eduarda Carretero Brezovsky , Mariana Maronezzi Ferraiol, Thiago Fanelli 2024-01-13T13:45:46Zoai:ojs.emnuvens.com.br:article/898Revistahttps://rebeca.socine.org.br/1ONGhttps://rebeca.socine.org.br/1/oairebeca@socine.org.br2316-92302316-9230opendoar:2024-01-13T13:45:46Rebeca (São Paulo) - Sociedade Brasileira de Estudos de Cinema e Audiovisual (SOCINE)false |
dc.title.none.fl_str_mv |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described Algoritmos de recomendación: un estudio sobre cómo funciona y cómo se describe el sistema de recomendaciones de Netflix Algoritmos de recomendação: um estudo sobre como funciona e como é descrito o sistema de recomendações da Netflix |
title |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described |
spellingShingle |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described Rodrigues Lucena, Tiago Franklin Sistema de Recomendação Dados Plataforma de Streaming Algoritmos Streaming Netflix Algoritmos de recomendação tecnologia Recommender Systems Data Streaming Platform Algorithms Recommender Systems Data Streaming Platform Algorithms |
title_short |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described |
title_full |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described |
title_fullStr |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described |
title_full_unstemmed |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described |
title_sort |
Recommendation Algorithms: a study on how Netflix’s recommender system works, and it is described |
author |
Rodrigues Lucena, Tiago Franklin |
author_facet |
Rodrigues Lucena, Tiago Franklin Garcia, Eduarda Carretero Brezovsky , Mariana Maronezzi Ferraiol, Thiago Fanelli |
author_role |
author |
author2 |
Garcia, Eduarda Carretero Brezovsky , Mariana Maronezzi Ferraiol, Thiago Fanelli |
author2_role |
author author author |
dc.contributor.author.fl_str_mv |
Rodrigues Lucena, Tiago Franklin Garcia, Eduarda Carretero Brezovsky , Mariana Maronezzi Ferraiol, Thiago Fanelli |
dc.subject.por.fl_str_mv |
Sistema de Recomendação Dados Plataforma de Streaming Algoritmos Streaming Netflix Algoritmos de recomendação tecnologia Recommender Systems Data Streaming Platform Algorithms Recommender Systems Data Streaming Platform Algorithms |
topic |
Sistema de Recomendação Dados Plataforma de Streaming Algoritmos Streaming Netflix Algoritmos de recomendação tecnologia Recommender Systems Data Streaming Platform Algorithms Recommender Systems Data Streaming Platform Algorithms |
description |
The audiovisual productions watched by each user on Netflix – a streaming platform - are based, in part, on data collection, computing, and archiving about how and what was previously consumed by each user and by others. Suggestions for new content are made by recommendation systems and operated by a set of algorithms, which are often kept in commercial secret. Netflix, on its website, proposes a “high-level description” of its recommendation system “in layman's language.” This article analyzes how this text explains the functioning of these tools, articulating it with authors who were already part of the group of platform’s developers, other critics, and specialists in algorithms. The analysis demonstrated that from the collection of little user data, especially if compared with the quantity of data usually extracted from social networking sites, it is possible to implement the recommendation system in an elaborated and customized way. The collected data behave as an inclusion pattern and constitute the raw material of a database that feeds the system, creating a complex personalized profile for each user. This profile is what recommends new titles in the search system and mainly guides the item’s position in the ranks in the initial interface. Finally, the position of the title in the interface and the row in which it is displayed significantly influence the choice of production. This, in turn, has consequences in the user’s contact with the diversity of audiovisual productions, the maintenance of the subscription, and the consumption experience on the platform. |
publishDate |
2024 |
dc.date.none.fl_str_mv |
2024-01-13 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Artigo avaliado pelos pares Texto info:eu-repo/semantics/other |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://rebeca.socine.org.br/1/article/view/898 10.22475/rebeca.v12n2.898 |
url |
https://rebeca.socine.org.br/1/article/view/898 |
identifier_str_mv |
10.22475/rebeca.v12n2.898 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.relation.none.fl_str_mv |
https://rebeca.socine.org.br/1/article/view/898/582 |
dc.rights.driver.fl_str_mv |
http://creativecommons.org/licenses/by/4.0 info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
http://creativecommons.org/licenses/by/4.0 |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Socine - Sociedade Brasileira de Estudos de Cinema e Audiovisual |
publisher.none.fl_str_mv |
Socine - Sociedade Brasileira de Estudos de Cinema e Audiovisual |
dc.source.none.fl_str_mv |
Rebeca - Brazilian Journal for Cinema and Audiovisual Studies; Vol. 12 No. 2 (2023): Rebeca 24 Rebeca - Revista Brasileira de Estudos de Cinema e Audiovisual; Vol. 12 Núm. 2 (2023): Rebeca 24 Rebeca - Revista Brasileira de Estudos de Cinema e Audiovisual; Vol. 12 No 2 (2023): Rebeca 24 Rebeca - Revista Brasileira de Estudos de Cinema e Audiovisual; v. 12 n. 2 (2023): Rebeca 24 2316-9230 reponame:Rebeca (São Paulo) instname:Sociedade Brasileira de Estudos de Cinema e Audiovisual (SOCINE) instacron:SOCINE |
instname_str |
Sociedade Brasileira de Estudos de Cinema e Audiovisual (SOCINE) |
instacron_str |
SOCINE |
institution |
SOCINE |
reponame_str |
Rebeca (São Paulo) |
collection |
Rebeca (São Paulo) |
repository.name.fl_str_mv |
Rebeca (São Paulo) - Sociedade Brasileira de Estudos de Cinema e Audiovisual (SOCINE) |
repository.mail.fl_str_mv |
rebeca@socine.org.br |
_version_ |
1809730937891586048 |