Global Collaboration in Artificial Intelligence: Bibliometrics and Network Analysis from 1985 to 2019
Haotian Hu,Dongbo Wang,Sanhong Deng
DOI: https://doi.org/10.2478/jdis-2020-0027
2020-01-01
Abstract:Abstract Purpose This study aims to explore the trend and status of international collaboration in the field of artificial intelligence (AI) and to understand the hot topics, core groups, and major collaboration patterns in global AI research. Design/methodology/approach We selected 38,224 papers in the field of AI from 1985 to 2019 in the core collection database of Web of Science (WoS) and studied international collaboration from the perspectives of authors, institutions, and countries through bibliometric analysis and social network analysis. Findings The bibliometric results show that in the field of AI, the number of published papers is increasing every year, and 84.8% of them are cooperative papers. Collaboration with more than three authors, collaboration between two countries and collaboration within institutions are the three main levels of collaboration patterns. Through social network analysis, this study found that the US, the UK, France, and Spain led global collaboration research in the field of AI at the country level, while Vietnam, Saudi Arabia, and United Arab Emirates had a high degree of international participation. Collaboration at the institution level reflects obvious regional and economic characteristics. There are the Developing Countries Institution Collaboration Group led by Iran, China, and Vietnam, as well as the Developed Countries Institution Collaboration Group led by the US, Canada, the UK. Also, the Chinese Academy of Sciences (China) plays an important, pivotal role in connecting the these institutional collaboration groups. Research limitations First, participant contributions in international collaboration may have varied, but in our research they are viewed equally when building collaboration networks. Second, although the edge weight in the collaboration network is considered, it is only used to help reduce the network and does not reflect the strength of collaboration. Practical implications The findings fill the current shortage of research on international collaboration in AI. They will help inform scientists and policy makers about the future of AI research. Originality/value This work is the longest to date regarding international collaboration in the field of AI. This research explores the evolution, future trends, and major collaboration patterns of international collaboration in the field of AI over the past 35 years. It also reveals the leading countries, core groups, and characteristics of collaboration in the field of AI.