Automated Narratives and Journalistic Text Generation: The Lead Organization Structure Translated into Code.

Brazilian Journalism Research

Endereço:
Faculdade de Comunicação, Universidade de Brasília(UnB), ICC Norte, Subsolo, Sala ASS 633
Brasília / DF
70910900
Site: http://bjr.sbpjor.org.br/index.php/bjr
Telefone: (61) 3307-6541
ISSN: 1981-9854
Editor Chefe: Fábio Henrique Pereira
Início Publicação: 31/05/2005
Periodicidade: Quadrimestral
Área de Estudo: Comunicação

Automated Narratives and Journalistic Text Generation: The Lead Organization Structure Translated into Code.

Ano: 2016 | Volume: 12 | Número: 1
Autores: Márcio Carneiro dos Santos
Autor Correspondente: Márcio Carneiro dos Santos | [email protected]

Palavras-chave: automated narratives, online journalism, Python, artificial intelligence, NLTK

Resumos Cadastrados

Resumo Inglês:

It describes the experiment of building a software capable of generating leads and newspaper titles in an automated fashion from information obtained from the Internet. The theoretical possibility Lage already provided by the end of last century is based on relatively rigid and simple structure of this type of story construction, which facilitates the representation or translation of its syntax in terms of instructions that the computer can execute. The paper also discusses the relationship between society, technique and technology, making a brief history of the introduction of digital solutions in newsrooms and their impacts. The development was done with the Python programming language and NLTK- Natural Language Toolkit library - and used the results of the Brazilian Soccer Championship 2013 published on an internet portal as a data source.