Through two experiments , three methods for constructing
<term>
word vectors
</term>
, i.e. ,
<term>
LSA-based , cooccurrence-based and dictionary-based methods
</term>
, were compared in terms of the ability to represent two kinds of
<term>
similarity
</term>
, i.e. ,
<term>
taxonomic similarity
</term>
and
<term>
associative similarity
</term>
.
#12444Through 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.