Journal article
Overlaying communities and topics: an analysis on publication networks
Scientometrics, v 90(2), pp 499-513
Feb 2012
Abstract
Two layers of enriched information are constructed for communities: a paper-to-paper network based on shared author relations and a paper-to-paper network based on shared word relations. k-means and VOSviewer, a modularity-based clustering technique, are used to identify publication clusters in the two networks. Results show that a few research topics such as webometrics, bibliometric laws, and language processing, form their own research community; while other research topics contain different research communities, which may be caused by physical distance.
Metrics
Details
- Title
- Overlaying communities and topics: an analysis on publication networks
- Creators
- Erjia Yan - School of Library and Information Science Indiana University Bloomington IL USAYing Ding - School of Library and Information Science Indiana University Bloomington IL USAElin Jacob - School of Library and Information Science Indiana University Bloomington IL USA
- Publication Details
- Scientometrics, v 90(2), pp 499-513
- Publisher
- Springer Netherlands; Dordrecht
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Information Science
- Web of Science ID
- WOS:000299088900010
- Scopus ID
- 2-s2.0-84855538524
- Other Identifier
- 991014976884604721
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Web of Science research areas
- Computer Science, Interdisciplinary Applications
- Information Science & Library Science