The impact of collaboration and knowledge networks on citations

Research papers not only involve author collaboration networks but also relate to knowledge networks. Previous research claims that a paper's citations are related to the node attributes of its authors in collaboration networks. We further propose that a paper's citations can also be affected by the node attributes of its knowledge elements in knowledge networks. In this study, we develop a new method to construct the knowledge network using article keywords. Further, we explore the antecedents of paper citations from both the collaboration and knowledge network perspectives. Using wind energy paper data (16,351 records) collected from WoS (Web of Science) and JCR (Journal Citation Reports) database, we construct two distinct networks and empirically examine the hypotheses of the relationships between node attributes of two networks and the paper's citations, which fill the gap in prior studies and will inspire related studies. We have the following key findings: in the collaboration network, the structural holes of authors have positive but non-significant effects on the paper's citations, while the authors' centrality has inverted U effects on the paper's citations; in the knowledge network, the structural holes of knowledge elements are positively related to the paper's citations, and the knowledge elements centrality has an inverted U relationship with the paper's citations.