rundimeco / text-similarity

text-similiarity

First version.

Usage :

python Compute_dist.py DIR_WITH_TEXTS

for instance :

python Compute_dist.py tmp/

Or, if you have a json (see format below): python Compute_dist.py JSON_FILE

• If launched as a script:
  • can take a path to a directory as parameter, will compute distances for all files inside
  • can also take a json file as parameter (see below)
• You can also launch the get_dist function with a dictionnary as parameter (see below)

The Json/dictionnary format

The entry format is a dictionary with two keys:

• "texts" : a list of texts
• "names" : their name or identifiers (useful for visualisation)

Example :

{ "texts":[ "toto", "titi", "tata" ], "names":[ "Text a", "Text b", "Text c", ] }