Download & Setups


BiblioTools scripts have been written in Python, and the codes should run in the majoruty of OS (tests have been conducted on Windows, MacOS and Linux). The BiblioMaps visualizations have been written in Javascript, using in particular the d3.js libraries. They should run in most modern browsers (tests have been made on Chrome, Mozilla and Safari).

In order to run BiblioTools, you'll need to have Python installed on your system. While the scripts should be compatible with both Python 2 and Python 3, we recommand that you use Python 3, which will be primarily used in future updates. To install Python on your system, you may download it from You'll need to check that the numpy, argparse, json, itertools, unidecode, collections, subprocess and networkx packages are installed. Some of these packages may already be installed with Python 3. If not, here's a tutorial on how to install python packages you may refer to.

pip install numpy
pip install argparse
pip install json
pip install itertools
pip install unidecode
pip install collections
pip install subprocess
pip install networkx

Depending on what you want to do, you might also need:

  • an access to either the Web of Science or Scopus, to extract bibliographic data to analyse.
  • a simple text editor, such as Sublime Text, to write some corrections in .txt, .html or .json files.
  • a localhost set up on your system, if you want to use the BiblioMaps interface locally. Here's a link to webpage explaining how to set up a local web server on most OS.
  • the Gephi software, to build network visualizations.
  • a latex editor to obtain automatic pdf reports of some of the analysis. Texmaker is a good one, working on most OS.


Finally, you may download and unzip the BiblioTools 3.2 repository file, and place the unzipped folder where it's most convenient for you.

Refer to the tutorials for a detailed description on how to use the software.


There are no publication devoted to the presentation of the BiblioMaps interface (yet), but if you use the tools, please refer to the BiblioTools-assiated publication: S Grauwin, P Jensen, Mapping Scientific Institutions. Scientometrics 89(3), 943-954 (2011).

Part of the credit should also go to Isabelle Sperano, who helped me improve the design of the BiblioMap interface.

Brief history of the project

The python BiblioTools scripts were primarily written during my post-doc at Inria in 2012. I developped the BiblioMaps web interface in 2016-2017, while working on free-lance projects.

Mars 2018Update of the python scripts to make them compatible with both the 1.X and the 2.0 versions of the networkx package (some of the 1.X method were depreciated in the new version).
February 2018Release of the BiblioTools 3.2 scripts, which include a few new functionalities (handling of both WOS and Scopus data, filtering options, automatic generation of analysis on successive time periods, etc) and above all the codes of the BiblioMaps interface, giving access to a series of interactive visualizaions (corpus description, maps of science, BC networks, etc).
2014-2017Development of a few versions with added/modified functionalities. These versions have not been publicly released here, but may have circulated among some people.
February 2014Small update of the BiblioTools 2.2 scripts to account for a change in the WOS exported data format.
September 2012Release of BiblioTools 2.2, updating the previous version to correct a few bugs and significantly improve the running time of some scripts.
July 2012Release of BiblioTools 2.1, a set of python scripts allowing to compute frequency distributions of bibliographic items in a given corpus and prepare co-occurrence networks and bibliographic coupling networks to visualize on Gephi.
Summer 2011Sharing of BiblioTools 1.0, a set of SQL queries used to perform a few analysis on WOS bibliographic data.