/EUSN-co-citation-networks

R code to get co-citation networks on social networks in the social sciences vs physics and computer science using Web of Science data.

Primary LanguageR

Social network and network science co-citations across disciplines in 1996-2013

Intro

  • This is an R project to create co-citation networks for the field of social network analysis and network science using Web of Science (WOS) data.
  • The project was originally created for EUSN Visualization Award 2014. It won the award in June 2014.
  • The project uses co-citation networks to map the research field of social networks and network science across different disciplines, namely the social sciences, physics, and computer science. Each node is a cited author and each tie is a co-citation between two authors. The citing papers are all the publications in the Thomson Reuters Web of Science that mention "social networks" or "network science" in their topic.
  • Using WOS categories, citing papers have been divided into two broad disciplines: the social sciences versus physics and computer science.
  • Citing papers have been downloaded for the period from 1996 to 2013, and divided into six 3-years time spans according to their publication date. This has resulted into a time series of six co-citation networks. A network animation has been created to show the evolution of the co-citation network over these six time periods.
  • The project uses the R packages igraph, network, networkDynamic and ndtv for network analysis and for producing network animations.

Data

  • All the original data as downloaded from the WOS are in ./Data/ in txt format.
  • The WOS search that produced the data is as follows:
TOPIC: ("social network" OR "social networks" OR "network science")
Timespan: 2011-2013. Indexes: SCI-EXPANDED, SSCI, A&HCI.
  • In the search above, 2011-2013 was set as the publication timespan. The same search was repeated for 5 more timespans: 1996-1998, 1999-2001, 2002-2004, 2005-2007, 2008-2010.

Code

  • All the R code needed to extract the co-citation networks from the txt data files and to create the static (png) and dynamic (mp4) figures is in ./Script/.
  • All the code can be run from 00_main.R, which sources the other R files included in the ./Script/ folder.
  • Note that 00_main.R is also where you set the relevant path for your working directory: you have to change that path for the project to be reproducible on your computer.

Figures and animations

  • All the figure and animation files are in ./Figures/.
  • In the figures (png), node size represents degree centrality, and edge width and color intensity represent the number of co-citations between two authors.
  • Node color represents the discipline in which an author is cited: (1) Blue authors are cited by papers in the social sciences; (2) Grey authors are cited by papers in physics or computer science; (3) Red authors are cited by papers in both the disciplinary classes.