A produção científica sobre inteligência artificial e seus impactos: análise de indicadores bibliométricos e altmétricos

• Ano de defesa: 2020
• Autor(a) principal: Marília Catarina Andrade Gontijo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ECI - ESCOLA DE CIENCIA DA INFORMAÇÃO; Programa de Pós-Graduação em Gestão e Organização do Conhecimento; UFMG
• Palavras-chave em Português: Comunicação científica; Estudos métricos da informação; Bibliometria; Altmetria; Inteligência Artificial.; Ciência da informação; Comunicação na ciência; Inteligência artificial
• Link de acesso: http://hdl.handle.net/1843/33913

The aim of this research was to analyze the performance of scientific production on artificial intelligence using bibliometrics and altmetrics indicators. These complementary indicators allow mapping literatures from many fields and areas of knowledge, analyzing their academic and social impact, verifying how scientific production is treated within academic communities, and what level of attention and visibility are displayed in Social Web sources. The motivation for the Artificial Intelligence theme arose with the growth of researches and studies from the field and the interest of several sectors of society for it. For the theoretical foundation, the research approached themes about topics on scientific communication, metric studies of information: bibliometrics and altmetrics, and their indicators, such as the artificial intelligence theme, its applications and its public and social interest. The research configured as descriptive, exploratory, bibliographic and of quantitative method, with the application of bibliometrics and altmetrics, which allowed analyzing academic performance, through citations, and social performance, through the on-line attention to scientific production on artificial intelligence in open access. A collection of the universe of research and bibliometrics and citation data was carried out in the Dimensions database, between the years 2012 and 2019. In order to collect altmetrics data, the Altmetric.com system, through the Altmetric Explorer, was used. The scientific production is analyzed in its distribution by date of publication, countries, institutions, scientific journals and fields of research, by the impact of citing authors and scientific articles and their visibility in the sources of the Social Web. For academic performance, the results show that, of the 7,453 articles retrieved from the universe, 3,994 (53.58%) of them obtained one or more citations, receiving 40,181 citations in total, considered as a moderate coverage of citations; while their social performance was considered low, with 2,411 (32.34%) of the publications receiving some kind of on-line attention, such as mentions in the sources of the Social Web, with the grand total of 120,746 mentions. There was an interdisciplinary approach to the topic, as well as countries and regions that have the greatest influence both academically and socially, such as the United States and the United Kingdom, and the most used Social Web sources, Twitter, Facebook and news. It was allowed to perceive the complementarity of bibliometrics and altmetrics in application, which mapped the scientific production in the field of Artificial Intelligence, verifying growth trends and patterns of research, citations and mentions in the sources of the Social Web.