Department of Computer Engineering, Faculty of Engineering, University of Kurdistan, Sanandaj, Iran
Abstract
The analysis of co-authorship networks serves as a key tool for identifying the structure of scientific collaborations. In this study, we examine these networks by employing advanced data mining and network analysis techniques on a comprehensive dataset of 2,345 articles extracted from PubMed, Scopus, and Semantic Scholar. By evaluating node centrality, clustering (using the Louvain algorithm), and density analysis of collaboration trends over the past two decades, our investigation provides a detailed picture of the evolution of scientific partnerships. Our results reveal a significant expansion of interdisciplinary collaborations, particularly in the fields of engineering, medicine, and computer science. The cluster analysis illuminates the role of bridging nodes in fostering knowledge development and strengthening interdisciplinary interactions. These insights not only deepen our understanding of the dynamics of scientific collaborations but also establish a framework for identifying research gaps and enhancing cooperative efforts. This study offers a comprehensive overview of co-authorship networks and suggests strategies for researchers and policymakers to reinforce scientific interactions and facilitate knowledge transfer. The innovation of our approach lies in the use of a large, refined dataset combined with temporal trend analysis and advanced network methods for delineating research communities, rendering this work a valuable reference for future studies.