Positive filter
taxonomic, similarity 2
(64.0 per million)
other,36-2-P06-2110,ak
of
<term>
similarity
</term>
, i.e. ,
<term>
taxonomic similarity
</term>
and
<term>
associative similarity
</term>
#12467Through two experiments, three methods for constructing word vectors, i.e., LSA-based, cooccurrence-based and dictionary-based methods, were compared in terms of the ability to represent two kinds of similarity, i.e., taxonomic similarity and associative similarity.
other,13-3-P06-2110,ak
word vectors
</term>
better reflect
<term>
taxonomic similarity
</term>
, while the
<term>
LSA-based and the
#12486The result of the comparison was that the dictionary-based word vectors better reflect taxonomic similarity , while the LSA-based and the cooccurrence-based word vectors better reflect associative similarity.