Positive filter
LSA-based, and, the, cooccurrence-based, word, vectors 1
(32.0 per million)
model,18-3-P06-2110,ak
The result of the comparison was that the
<term>
dictionary-based word vectors
</term>
better reflect
<term>
taxonomic similarity
</term>
, while the
<term>
LSA-based and the cooccurrence-based word vectors
</term>
better reflect
<term>
associative similarity
</term>
.
#12491The 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.