python - 在 Python 中获取新单词的派生形式

Question

我想知道我是否可以获得给定单词的所有派生形式

例如，给定单词“good”，我得到“goodness”和“advantage”等。

特别是，得到一个“形容词”的相关名词

谢谢

Answer 1

It's going to be somewhat difficult to do what you're wanting for a few reasons, especially the problem of ambiguity. This can be alleviated somewhat if you choose the correct sense of the word, or otherwise do automatic word sense disambiguation (WSD).

As an example, for the word 'good', there are 27 senses in WordNet, and 37 unique lemmas (dictionary forms).

Here is a quick example using NLTK's implementation of WordNet.

>>> from nltk.corpus import wordnet
>>> good = wordnet.synsets('good')

>>> lemmas = set()
>>> for synset in good:
    for lemma in synset.lemmas:
        lemmas.add(lemma.name)
>>> lemmas
set(['beneficial', 'right', 'secure', 'just', 'unspoilt', 'respectable', 'good', 'goodness', 'dear', 'salutary', 'ripe', 'expert', 'skillful', 'in_force', 'proficient', 'unspoiled', 'dependable', 'soundly', 'honorable', 'full', 'undecomposed', 'safe', 'adept', 'upright', 'trade_good', 'sound', 'in_effect', 'practiced', 'effective', 'commodity', 'estimable', 'well', 'honest', 'near', 'skilful', 'thoroughly', 'serious'])
>>> len(lemmas)
37

>>> for synset in good:
    print synset
    print synset.lemmas
    print '-' * 79

Synset('good.n.01')
[Lemma('good.n.01.good')]
-------------------------------------------------------------------------------
Synset('good.n.02')
[Lemma('good.n.02.good'), Lemma('good.n.02.goodness')]
-------------------------------------------------------------------------------
Synset('good.n.03')
[Lemma('good.n.03.good'), Lemma('good.n.03.goodness')]
-------------------------------------------------------------------------------
Synset('commodity.n.01')
[Lemma('commodity.n.01.commodity'), Lemma('commodity.n.01.trade_good'), Lemma('commodity.n.01.good')]
-------------------------------------------------------------------------------
Synset('good.a.01')
[Lemma('good.a.01.good')]
-------------------------------------------------------------------------------
Synset('full.s.06')
[Lemma('full.s.06.full'), Lemma('full.s.06.good')]
-------------------------------------------------------------------------------
Synset('good.a.03')
[Lemma('good.a.03.good')]
-------------------------------------------------------------------------------
Synset('estimable.s.02')
[Lemma('estimable.s.02.estimable'), Lemma('estimable.s.02.good'), Lemma('estimable.s.02.honorable'), Lemma('estimable.s.02.respectable')]
-------------------------------------------------------------------------------
Synset('beneficial.s.01')
[Lemma('beneficial.s.01.beneficial'), Lemma('beneficial.s.01.good')]
-------------------------------------------------------------------------------
Synset('good.s.06')
[Lemma('good.s.06.good')]
-------------------------------------------------------------------------------
Synset('good.s.07')
[Lemma('good.s.07.good'), Lemma('good.s.07.just'), Lemma('good.s.07.upright')]
-------------------------------------------------------------------------------
Synset('adept.s.01')
[Lemma('adept.s.01.adept'), Lemma('adept.s.01.expert'), Lemma('adept.s.01.good'), Lemma('adept.s.01.practiced'), Lemma('adept.s.01.proficient'), Lemma('adept.s.01.skillful'), Lemma('adept.s.01.skilful')]
-------------------------------------------------------------------------------
Synset('good.s.09')
[Lemma('good.s.09.good')]
-------------------------------------------------------------------------------
Synset('dear.s.02')
[Lemma('dear.s.02.dear'), Lemma('dear.s.02.good'), Lemma('dear.s.02.near')]
-------------------------------------------------------------------------------
Synset('dependable.s.04')
[Lemma('dependable.s.04.dependable'), Lemma('dependable.s.04.good'), Lemma('dependable.s.04.safe'), Lemma('dependable.s.04.secure')]
-------------------------------------------------------------------------------
Synset('good.s.12')
[Lemma('good.s.12.good'), Lemma('good.s.12.right'), Lemma('good.s.12.ripe')]
-------------------------------------------------------------------------------
Synset('good.s.13')
[Lemma('good.s.13.good'), Lemma('good.s.13.well')]
-------------------------------------------------------------------------------
Synset('effective.s.04')
[Lemma('effective.s.04.effective'), Lemma('effective.s.04.good'), Lemma('effective.s.04.in_effect'), Lemma('effective.s.04.in_force')]
-------------------------------------------------------------------------------
Synset('good.s.15')
[Lemma('good.s.15.good')]
-------------------------------------------------------------------------------
Synset('good.s.16')
[Lemma('good.s.16.good'), Lemma('good.s.16.serious')]
-------------------------------------------------------------------------------
Synset('good.s.17')
[Lemma('good.s.17.good'), Lemma('good.s.17.sound')]
-------------------------------------------------------------------------------
Synset('good.s.18')
[Lemma('good.s.18.good'), Lemma('good.s.18.salutary')]
-------------------------------------------------------------------------------
Synset('good.s.19')
[Lemma('good.s.19.good'), Lemma('good.s.19.honest')]
-------------------------------------------------------------------------------
Synset('good.s.20')
[Lemma('good.s.20.good'), Lemma('good.s.20.undecomposed'), Lemma('good.s.20.unspoiled'), Lemma('good.s.20.unspoilt')]
-------------------------------------------------------------------------------
Synset('good.s.21')
[Lemma('good.s.21.good')]
-------------------------------------------------------------------------------
Synset('well.r.01')
[Lemma('well.r.01.well'), Lemma('well.r.01.good')]
-------------------------------------------------------------------------------
Synset('thoroughly.r.02')
[Lemma('thoroughly.r.02.thoroughly'), Lemma('thoroughly.r.02.soundly'), Lemma('thoroughly.r.02.good')]
-------------------------------------------------------------------------------

Answer 2

我建议查看 NLTK 中的WordNet 语料库。更多关于 WordNet的信息。