Skip to content

Latest commit

 

History

History
90 lines (74 loc) · 4.44 KB

README.md

File metadata and controls

90 lines (74 loc) · 4.44 KB
Raw

Similarity Network of first names given in Belgium between 1995 and 2015

Data source

Belgian Federal Public Service for Economy, SMEs, Self-Employed and Energy:

Download these *.xls files in your local data folder, next to the src folder

Usage

$ python src/build_firstname_network.py -h

usage: build_firstname_network.py [-h] --inFileXls INFILEXLS [INFILEXLS ...]
                                  [--sheetName SHEETNAME]
                                  [--partOfCountry PARTOFCOUNTRY]
                                  [--startNumber STARTNUMBER]
                                  [--maxNrNames MAXNRNAMES]
                                  [--simThreshold SIMTHRESHOLD]
                                  [--degreeThreshold DEGREETHRESHOLD]
                                  [--rankThreshold RANKTHRESHOLD]
                                  [--bonusMultiplier BONUSMULTIPLIER]
                                  --outFileGraphML OUTFILEGRAPHML

optional arguments:
  -h, --help            show this help message and exit
  --inFileXls INFILEXLS [INFILEXLS ...]
                        input file(s) with Belgian first names in MS Excel
                        format
  --sheetName SHEETNAME
                        name of the sheet - i.e. year - of interest, e.g. 2000
                        (default is 1995 through 2015)
  --partOfCountry PARTOFCOUNTRY
                        1 = whole of Belgium (DEFAULT); 2 = Brussels only;
                        3=Flanders only; 4=Wallonia only)
  --startNumber STARTNUMBER
                        rank number of the highest-ranked first name to
                        include in output graph
  --maxNrNames MAXNRNAMES
                        number of names to store in output network
  --simThreshold SIMTHRESHOLD
                        minimum inter-name similarity for a link to be created
  --degreeThreshold DEGREETHRESHOLD
                        minimum degree required for a node to be included in
                        the output graph
  --rankThreshold RANKTHRESHOLD
                        all nodes below this rank are guaranteed to be
                        included in the output graph
  --bonusMultiplier BONUSMULTIPLIER
                        the edge weights of the nodes below rankThreshold get
                        multiplied with this bonus to increase their chances
                        of survival
  --outFileGraphML OUTFILEGRAPHML
                        output file with Belgian first names in GraphML format

How the GraphML files were built

$ mkdir out
$ python src/build_firstname_network.py --inFileXls data/Voornamen_Jongens_1995-2015_tcm325-239464.xls \\
                                        --outFileGraphML out/firstname.graphml \\
                                        --startNumber 1 --maxNrNames 3751 \\
                                        --simThreshold 0.55 \\
                                        --degreeThreshold 2 \\
                                        --rankThreshold 100

$ python src/build_firstname_network.py --inFileXls data/Voornamen_meisjes_1995-2015_tcm325-239448.xls \\
                                        --outFileGraphML out/firstname.graphml \\
                                        --startNumber 1 \\
                                        --maxNrNames 4135 \\
                                        --simThreshold 0.55 \\
                                        --degreeThreshold 2 \\
                                        --rankThreshold 100

Visualization with Gephi

  • Load the produced GraphML file into Gephi
  • Color the nodes via Appearance - Nodes - Attributes - Choose an attribute - community
  • Arrange the nodes via Layout - Fruchterman Reingold with the following settings:
    • Area: 100000.0
    • Gravity: 0.5
    • Speed: 10.0
  • Export to PDFs via Preview - Export PDF
  • Export to interactive web page via File - Export - Sigma.js template

Result files and full story

See my blog article