Percolation and connectivity formation in the dynamics of data citation networks in high energy physics
Objectives. The object of the research is to study citation information networks structured on the basis of a sample from the arXiv database related to theoretical high energy physics (high energy physics, HEP). Since 1974, this database has indexed more than 500000 articles, including their complet...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | Russian |
| Published: |
MIREA - Russian Technological University
2025-02-01
|
| Series: | Российский технологический журнал |
| Subjects: | |
| Online Access: | https://www.rtj-mirea.ru/jour/article/view/1064 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Objectives. The object of the research is to study citation information networks structured on the basis of a sample from the arXiv database related to theoretical high energy physics (high energy physics, HEP). Since 1974, this database has indexed more than 500000 articles, including their complete citation trees. The paper proposes a method for detecting percolation transitions in the dynamics of cluster formation of articles with similar content. Improving the accuracy of information cycles in knowledge networks can help resolve applied problems related to the quality of scientometrics and its indicators.Methods. An optimized algorithm for dynamic network separation in the Pajek software environment was applied, in order to detect the emergence of a largest component equivalent to a percolation transition. This approach enables a detailed study of dynamic and general parameters to be carried out in each reduced network with a given time step. The clustering algorithm combines citation structure and temporal information about data.Results. It was found that a percolation transition occurs in the HEP network. The indicator of this transition is the formation of a largest component near the critical point which occurs at the 10th month of the time sample interval. At the same time, a generalized conclusion about the behavior of network parameters shows a positive trend in the growth of connectivity for the entire time period (from 1991 to 2003). Furthermore, a generalized analysis of citation distribution reveals eleven laureates of highly cited articles who set the basic vector for development in the field of HEP. It is worth noting that the prominent scientists from the top three in terms of citations are linked by a shared field of research: string theory. Verification of this fact confirms that our citation evaluation method is effective. Determining the characteristics of the HEP (high-energy physics) network enables an important indicator of the researcher’s activity and behavior to be identified.Conclusions. In the column of authors linked by co-authorship, of the 9200 authors in the HEP physics community, 7304 belong to a single connected component. The temporal nature of citations indicates a rapid uptake and understanding of relevant new work. Percolation transitions, which are indicators of sudden conceptual shifts in citation networks, allow us to identify and link articles into research schemes which form clusters of new ideas and theories. |
|---|---|
| ISSN: | 2782-3210 2500-316X |