|
gptkbp:instanceOf
|
gptkb:convolutional_neural_network
word embedding technique
|
|
gptkbp:application
|
gptkb:machine_learning
information retrieval
machine translation
natural language processing
sentiment analysis
text classification
semantic similarity
|
|
gptkbp:architecture
|
gptkb:Continuous_Bag_of_Words_(CBOW)
Skip-gram
|
|
gptkbp:citation
|
high
|
|
gptkbp:developedBy
|
gptkb:Google
gptkb:Tomas_Mikolov
|
|
gptkbp:dimensions
|
configurable
|
|
gptkbp:feature
|
low memory usage
efficient training
captures semantic relationships
captures syntactic relationships
predicts context words
predicts target word
scalable to large datasets
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Word2Vec
|
|
gptkbp:influenced
|
gptkb:ELMo
gptkb:GloVe
gptkb:BERT
gptkb:FastText
|
|
gptkbp:input
|
gptkb:text
|
|
gptkbp:introducedIn
|
2013
|
|
gptkbp:language
|
gptkb:Python
gptkb:C++
|
|
gptkbp:license
|
gptkb:Apache_License_2.0
|
|
gptkbp:lossFunction
|
hierarchical softmax
negative sampling
|
|
gptkbp:notablePublication
|
gptkb:Efficient_Estimation_of_Word_Representations_in_Vector_Space
gptkb:Distributed_Representations_of_Words_and_Phrases_and_their_Compositionality
|
|
gptkbp:openSource
|
true
|
|
gptkbp:output
|
vector representations
|
|
gptkbp:purpose
|
learn word embeddings
|
|
gptkbp:relatedTo
|
gptkb:Gensim
gptkb:GloVe
gptkb:FastText
word embeddings
|
|
gptkbp:trainer
|
unsupervised learning
|
|
gptkbp:url
|
https://arxiv.org/abs/1301.3781
https://arxiv.org/abs/1310.4546
https://code.google.com/archive/p/word2vec/
|
|
gptkbp:usedIn
|
gptkb:Google_News_corpus
gptkb:Wikipedia_corpus
research
|
|
gptkbp:vectorOperation
|
king - man + woman = queen
|
|
gptkbp:bfsParent
|
gptkb:Gensim
gptkb:Google_Research
|
|
gptkbp:bfsLayer
|
6
|