<p>来自<a href="http://www.nltk.org/howto/wordnet.html" rel="nofollow">this</a></p>
<p><strong>路径相似度、wup_相似度和lch_相似度,所有这些都应该有效,因为它们基于Wordnet层次结构中两个synset之间的距离。</p>
<pre><code>dog = wn.synset('dog.n.01')
cat = wn.synset('cat.n.01')
dog.path_similarity(cat)
dog.lch_similarity(cat)
dog.wup_similarity(cat)
</code></pre>
<p/><hr/>
来自同一链接(相关部分以粗体显示)
<p><strong>synset1.路径相似性(synset2):</strong></p>
<blockquote>
<p>Return a score denoting how similar
two word senses are, <strong>based on the shortest path that connects the
senses in the is-a (hypernym/hypnoym) taxonomy</strong>. The score is in the
range 0 to 1, except in those cases where a path cannot be found (will
only be true for verbs as there are many distinct verb taxonomies), in
which case -1 is returned. A score of 1 represents identity i.e.
comparing a sense with itself will return 1.</p>
</blockquote>
<hr/>
<p><strong>synset1.lch_相似性(synset2),Leacock Chodorow相似性:</strong></p>
<blockquote>
<p>Return a
score denoting how similar two word senses are, <strong>based on the shortest
path that connects the senses</strong> (as above) and the maximum depth of the
taxonomy in which the senses occur. The relationship is given as
-log(p/2d) where p is the shortest path length and d the taxonomy depth.</p>
</blockquote>
<hr/>
<p><strong>synset1.wup_相似性(synset2),Wu-Palmer相似性:</strong></p>
<blockquote>
<p>Return a score
denoting how similar two word senses are, <strong>based on the depth of the
two senses in the taxonomy and that of their Least Common Subsumer
(most specific ancestor node)</strong>. Note that at this time the scores given
do <em>not</em> always agree with those given by Pedersen's Perl
implementation of Wordnet Similarity.</p>
</blockquote>