The Most-Cited Articles of the 21st Century

The aim of this paper is to collect the most-cited articles of the 21 century and to study how this group changed over time. Here the term “most-cited” is operationalized by considering yearly h-cores in the Web of Science. These h-cores are analysed in terms of authors, research areas, countries, institutions, journals and average number of authors per paper. We only consider publications of article or proceedings type. The research of some of the more prolific authors is on genetics and genomes publishing in multidisciplinary journals, such as Nature and Science, while the results show that writing a software tool for crystallography or molecular biology may help collecting large numbers of citations. English is the language of all articles in any h-core. The core institutions are largely those best placed in most rankings of world universities. Some attention is given on the relation between h-core articles and the information sciences. We conclude by stating that the notion of an h-core provides a new perspective on leading countries, articles and scientists. Conference Topic Citation and co-citation analysis Introduction The objective of this paper is to collect the most-cited articles of the 21 century and to study how this group changed over time. The term “most-cited” is operationalized by considering the h-core (Hirsch, 2005; Rousseau, 2006) in the Web of Science (WoS) for each period of time, starting with the period 2001-2005, continuing with 2001-2006 and ending with 20012013. These periods refer to the publication and the citation window. We recall that the h-core at a given moment in time, for instance on January 1, 2009, consists of the set of articles which at that time received a number of citations at least equal to their rank among all articles published during the period 2001-2008. This approach is different from the one taken in (Van Noorden et al., 2014) where a fixed number, concretely 100, of articles is considered. Furthermore, we study the papers making up the corresponding h-cores in terms of authors, research areas, countries, institutions, journals and average number of authors per paper. Methodology We have to point out that the 21 century starts on January 1, 2001. This implies that we only consider publications from 2001 on. Moreover, we only consider publications in Thomson Reuters’ Web of Science (WoS) and we restrict ourselves to publications of article or proceedings type. Although finding today’s h-core for a set of articles in the Web of Science is easy, finding an h-core in the past needs some specific knowledge of the tools available in the WoS. First one retrieves the set for which one wants to determine the h-core (ending in the year Y). Its