Python implementation of Lin similarity
Python
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
README.md
linsim.py

README.md

Lin Similarity

NAME

sim.py - computes Lin similarity of a given noun and all other nouns in a given file

SYNOPSIS

python sim.py INPUTFILE INPUTWORD

DESCRIPTION

sim.py is a simple program which computes the Lin similarity of a given input noun and all others nouns in the given input file, whereby similarity is defined as

the ratio between the amount of information in the commonality and the amount of information in the description of the two objects.

Dependency triples are extracted from the given input file and stored as features of the nouns. The amount of information contained in every single feature is calculated accordingly. Pairwise similarity is computed between the given input noun and nouns with at least one similar feature. The fifty most similar words are displayed in descending order of their similarity.

FILES

INPUTFILE

The input file must be in CoNLL09 format.

INPUTWORD

The input word must be a noun.

EXAMPLE

INPUTFILE

tiger_ release_ aug07.corrected.16012013.conll09

INPUTWORD

Mann

COMMAND

$ python sim.py tiger_ release_ aug07.corrected.16012013.conll09 Mann

OUTPUT

Mensch
Frau
Teil
Regierung
Million
Prozent
Land
Experte
Zahl
Präsident
...

AUTHOR

Melanie Tosik, tosik@uni-potsdam.de