The keyword to success: Comparative analysis of Computer Science research across representative universities in Latin America
- 1 April 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 99-105
Abstract
Scientific publications are often used to rank higher education institutions, countries, research centers, using parameters such as number of citations and relevance. At the same time, there is a lack of comparative information among the research conducted in Latin America. In this study, we aim to compare the Computer Science research in which Latin American scientists have participated, by analyzing the co-occurrence networks of keywords chosen for the corresponding scientific publications. To this aim, we collect information, from the Scopus website, about the scientific publications whose authors belong to one of the two best universities in eleven Latin American countries filtering scientific publications only in the area of Computer Science. Next, we build graphs per university, calculate some structural measures. Additionally, we define a metric named University Keyword Diversity (UKD) as a variation of the statistical equation of entropy for the keywords appearing in the university research articles. We correlate such metrics with the number of citations and the university ranking finding correlations as high as 0.8. Our results show that the university with the highest UKD in Latin America is Universidad Nacional Autónoma de México (UNAM) and that its corresponding graph is one of the richest in number of nodes and relations.Keywords
This publication has 11 references indexed in Scilit:
- Co-Occurrence Network of High-Frequency Words in the Bioinformatics Literature: Structural Characteristics and EvolutionApplied Sciences, 2018
- Multidimensional assessment of scholarly research impactJournal of the Association for Information Science and Technology, 2015
- Are Global Rankings Unfair to Latin American Universities?International Higher Education, 2015
- Co-authorship networks and research impact: A social capital perspectiveResearch Policy, 2013
- The Impact of Article Length on the Number of Future Citations: A Bibliometric Analysis of General Medicine JournalsPLOS ONE, 2013
- A combined bibliometric indicator to predict article impactInformation Processing & Management, 2011
- Changing the Culture of Science Education at Research UniversitiesScience, 2011
- Using content-based and bibliometric features for machine learning models to predict citation counts in the biomedical literatureScientometrics, 2010
- Modularity and community structure in networksProceedings of the National Academy of Sciences, 2006
- Collective dynamics of ‘small-world’ networksNature, 1998