Homepage  |  Publications  |   Software  |   Courseware; indicators  |   Animation  |   Geo  |   Search website (Google)

Fractional Counting of Bibliometric Networks: FNetwork.exe

 

FNetwork.exe reads a file with records downloaded from the Web-of-Science in the plain text format; that is, tagged with “AU” for authors, etc. For example:

 

FN Thomson Reuters Web of Science™

VR 1.0

PT J

AU Leydesdorff, L

             ….

 

The file has to be (re-)named “data.txt”. If more than one batch of 500 records was downloaded, one can use the following command from the command line (C:\[folder]\) to combine the files into a single one: 

             Copy saved*.* data.txt

 

FNetwork.exe first prompts for whether one wants the data to be organized relationally. This has to be done only the first time so that the relevant files are generated. Secondly, one is prompted for the level of aggregation: “a” for authors, “i” for institutes, or “c” for countries. Specify the units of analysis (vertices) of the network.

 

The program generates two files in the Pajek .net format (edgelist): mtrx.net and fmtrx.net. Mtrx.net contains the co-occurrence matrix; fmtrx1.net, fmtrx2.net or fmtrx3.net the fractionally counted co-occurrence matrix using:

 

                                                            u*_ij =                                                                (1)

                                                 u*_ij =  =                                                     (2)

                                         u*_ij =  =                                             (3)

Eq. 3 is the default; but one has the option to choose Eq. 1 or 2.

The background is explained in:

Leydesdorff, L., & Park, H. W. (2016). Full and Fractional Counting in Bibliometric Networks.

Perianes-Rodriguez, A., Waltman, L., & van Eck, N. J. (2016). Constructing bibliometric networks: A comparison between full and fractional counting. arXiv preprint arXiv:1607.02452.

  

See also:

Batagelj, V., & Cerinšek, M. (2013). On bibliographic networks. Scientometrics, 96(3), 845-864.

Newman, M. E. (2001). Scientific collaboration networks. II. Shortest paths, weighted networks, and centrality. Physical Review E, 64(1), 016132.

Park, H. W., Yoon, J., & Leydesdorff, L. (2016; in press). The Normalization of Co-authorship Networks in the Bibliometric Evaluation: The Government Stimulation Programs of China and Korea. Scientometrics arXiv preprint arXiv:1605.03593. doi: 10.1007/s11192-016-1978-2

 

Amsterdam, 9 October 2016.

 

A routine for fractional counting of authors and addresses at the paper level is available at http://www.leydesdorff.net/software/fraction/index.htm .

 

A further routine instfrac.exe produces the matrices for institutional collaboration: matrix.txt, matrix.dbf, fractionally counted, and on the basis of these matrices the network files coocc.dat and cosine.dat for processing in Pajek. See for further explanation at http://www.leydesdorff.net/software/instcoll  (September 23, 2015).

 

For institutional networks with more than 1,000 institutions, click here.  

 

Similarly, intfrac.exe produces the matrices for international collaboration: matrix.txt, matrix.dbf, fractionally counted, and on the basis of these matrices the network files coocc.dat and cosine.dat for processing in Pajek. See for further explanation at http://www.leydesdorff.net/software/intcoll (November 2, 2015)

 

November 2, 2015.