| dc.description | <p>Joint embeddings of selectional preferences, words, and fine-grained entity types.</p>
<p>The vocabulary consists of:
<ul>
<li> verbs and their dependency relation separated by "@", e.g. "sink@nsubj" or "elect@dobj"
<li> words and short noun phrases, e.g. "Titanic"
<li> fine-grained entity types using the FIGER inventory, e.g.: /product/ship or /person/politician
</ul>
</p>
<p>
The files are in word2vec binary format, which can be loaded in Python with gensim like this:
<code>
<pre>from gensim.models import KeyedVectors
emb_file = "/path/to/embedding_file"
emb = KeyedVectors.load_word2vec_format(emb_file, binary=True)</pre></code>
</p> | |
