Semiodiscursive analysis of TV newscasts based on data mining and image processing

• Autor(a) principal: Conceição, Felipe Leandro Andrade da
• Data de Publicação: 2017
• Outros Autores: Pádua, Flávio Luis Cardeal, Pereira, Adriano César Machado, Assis, Guilherme Tavares de, Silva, Giani David, Andrade, Antônio Augusto Braighi
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Acta scientiarum. Technology (Online)
• Texto Completo: http://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/article/view/29763
• topic: Journalism; Computing; Discursive Metadata; Discourse Analysis; Interdisciplinar

This work addresses the development of a novel computer-aided methodology for discourse analysis of TV newscasts. A TV newscast constitutes a particular type of discourse and has become a central part of the modern-day lives of millions of people. It is important to understand how this media content works and how it affects human life. To support the study of TV newscasts under the discourse analysis perspective, this work proposes a newscast structure to recover its main units and extract relevant data, named here as newscast discursive metadata (NDM). The NDM describes aspects, such as screen time and field size of newscasts’ participants and themes addressed. Data mining and image analysis methods are used to extract and analyze the NDM of a dataset containing 41 editions of two Brazilian newscasts. The experimental results are promising, demonstrating the effectiveness of the proposed methodology.