Language Levenshtein Distance

Python Script

• Uses Swadesh lists (sourced from Wiktionary) to calculate average Levenshtein distance between pairs of languages.
• To use, add Swadesh list as CSV file to swadesh_lists folder, and the language to the languages list in main.py.
• In addition, special character distances can be set using the following syntax in the CSV file:

  accept,{cost},{character 1},{character 2}
  

• The resulting output CSV will go into the output.csv file after running main.py, and this can be copied to the graphing page.

Graphing Webpage

• In /docs directory (for gh-pages)
• Uses D3.js to visualize average Levenshtein distances between languages.
• The file csv.js contains all of the options, including the output CSV from the Python script (csv), the list of languages (languageOptions), and language group definitions (languageGroups).
• Language group definitions are structured as follows:

  {
      name: '{language group name}',
      langs: ['{languages included}']
  }

• Hovering over a language in a graph will show the raw Levenshtein distances for all languages and the language being hovered over.
• The redder a language is, the greater its average Levenshtein distance with all others on the graph (this scale is relative to each graph).