|
چکیده
|
One area of growing interest in computational psychology is the analysis of psychopathological networks. Numerous related studies and several recent review articles have been published in this field. Understanding the characteristics, authors, relationships, and focus areas of the studies can provide greater benefits to researchers in this field. This article presents the first analysis of co-authorship networks in computational network-oriented psychopathology research. To this end, bibliographic data were collected from Google Scholar. Given the difficulty and potential for errors in manually reviewing the 2,799 research articles published between 2000 and 2022, co-authorship network analysis was conducted using machine learning methods for graph analysis. Network density, average degree, clustering coefficient, and the number of communities were calculated, and temporal changes were evaluated. Prominent authors were identified based on centrality measures. The co-authorship network for the entire period consisted of 6,025 nodes and 9,808 weighted edges. Time series analysis showed a linear correlation between the number of authors and the number of connections. Furthermore, the number of communities was linearly correlated with the number of authors. Identifying research clusters through topic modeling revealed that keywords such as user, event, family, and comments were the most commonly used representative texts in articles in this field. Additionally, we highlighted disorders that may have potential for more research in the field of network analysis, those with no related publications, for further investigation. Finally, the findings show a lack of collaboration between computer science researchers and specialists in this area.
|