Abstract
Understanding the nature and organization of scientific communities is of broad interest. The “Invisible College” is a historical metaphor for one such type of community that refers to a small group of scientists working on a problem of common interest. The scientific and social behavior of such colleges has been the subject of case studies that have examined limited samples of the scientific enterprise. We introduce a metamethod for large-scale discovery that consists of a pipeline to select themed article clusters, whose authors can then be analyzed. A sample of article clusters produced by this pipeline was reviewed by experts, who inferred significant thematic relatedness within clusters, suggesting that authors linked to such clusters may represent valid communities of practice. We explore properties of the author communities identified by our pipeline, and the publication and citation practices of both typical and highly influential authors. Our study reveals that popular domain-independent criteria for graphical cluster quality must be carefully interpreted in the context of searching for author communities, and also suggests a role for contextual criteria.
| Original language | American English |
|---|---|
| Pages (from-to) | 184-203 |
| Number of pages | 20 |
| Journal | The MIT Press Journals |
| Volume | 2 |
| Issue number | 1 |
| DOIs | |
| State | Published - 2020 |
Keywords
- Citation graph
- Clustering
- Community finding
- Invisible College
- Scientific organization
Fingerprint
Dive into the research topics of 'Finding scientific communities in citation graphs: Articles and authors: Articles and authors'. Together they form a unique fingerprint.Projects
- 1 Finished
-
Enhanced Research Network Informatics Environment (ERNIE)
Falk-Krzesinski, H. (PI), Chacko, G. (PI) & Aisati, M. (PI)
02/2/19 → 12/12/20
Project: Research
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver