sort of Natural language processing

Question

I have a word lets say "Racing".

I want to know all its formats nouns, adjectives verbs etc. For eg. Race racer races etc.

Is there a python library that can help me out with this?

Answer 1

if you're interested in matching these strings, I'd recommend looking at fuzzywuzzy. It allows you to easily tokenize strings and match based on token equivalence within a certain tolerance.

For example 'Race' and 'Racer' would have a high matching percentage of tokens and could be considered match.

As far as extrapolating different forms of a word, I'm afraid I've never seen something canned, although this could very well help toward your goal, and perhaps provide you the right starting place for writing your own.

I hope this helps

Answer 2

I found similar question Convert words between verb/noun/adjective forms

In one of the answers user @PBelzile suggests online service Idilia: http://www.idilia.com/demos/language-graph-browser/?node=code%2FV3

Looks like it can do what you want.

Also you can try nltk with wordnet:

import nltk
nltk.download("wordnet")

from nltk.corpus import wordnet as wn

for lemma in wn.lemmas("race"):
    for form in lemma.derivationally_related_forms():
        print lemma, form

Outputs:

Lemma('race.n.01.race') Lemma('race.v.02.race')
Lemma('race.n.02.race') Lemma('race.v.02.race')
Lemma('race.n.02.race') Lemma('racy.s.04.racy')
Lemma('race.n.03.race') Lemma('racial.a.01.racial')
Lemma('race.v.02.race') Lemma('racing.n.01.racing')
Lemma('race.v.02.race') Lemma('race.n.02.race')
Lemma('race.v.02.race') Lemma('racer.n.03.racer')
Lemma('race.v.02.race') Lemma('racer.n.02.racer')
Lemma('race.v.02.race') Lemma('race.n.01.race')
Lemma('race.v.04.race') Lemma('racer.n.01.racer')