Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
mind:word2vec [2016/09/25 16:23]
bayb2 created
mind:word2vec [2016/09/25 16:49] (current)
bayb2
Line 1: Line 1:
 =Word2Vec= =Word2Vec=
  
-==Notation== +==Intro== 
-Instead ​of saying+“Just as Van Gogh’s painting ​of sunflowers is a two-dimensional mixture of oil on canvas that represents vegetable matter in a three-dimensional space in Paris in the late 1880s, so 500 numbers arranged in a vector can represent a word or group of words.” --DL4J
  
- ​knee ​leg = elbow - arm+Word2Vec can guess a word’s association with other words, or cluster documents and define them by topic. It makes qualities into quantities, and similar things and ideas are shown to be “close” in its 500-dimension vectorspace.
  
-Or+Word2Vec is not classified as "deep learning"​ because it is only a 2-layer neural net.
  
- knee is to leg as elbow is to arm+Input -> text corpus 
 +Output -> set of vectors, or neural word embeddings
  
-Logical analogy notation is 
  
- knee : leg :: elbow : arm+===Examples===
  
 + Rome - Italy = Beijing - China, so Rome - Italy + China = Beijing
  
 + king : queen :: man : woman
  
-==Examples==+ house : roof :: castle : [dome, bell_tower, spire, crenellations,​ turrets]
  
 + China : Taiwan :: Russia : [Ukraine, Moscow, Moldova, Armenia]
 +
 +
 +==Notation==
 +===Algebraic notation===
 +
 + knee - leg = elbow - arm
 +
 +===English logic===
 +
 + knee is to leg as elbow is to arm
 +
 +===Logical analogy notation===
 +
 + knee : leg :: elbow : arm
  
-Rome - Italy Beijing ​Chinaso Rome - Italy + China = Beijing+==Models== 
 +===Continuous bag of words (CBOW) model===  
 +*Uses a context to predict a target word. Faster. 
 +*Several times faster to train than the skip-gramslightly better accuracy for frequent words.
  
-king : queen :: man : woman+===Skip-gram model===  
 +*Uses a word to predict a target context. 
 +*Works well with small amount of the training data, represents well even rare words or phrases. 
 +*Produces more accurate results on large datasets.
  
-house : roof :: castle : [domebell_tower, spire, crenellations,​ turrets]+==Implementation== 
 +Word2Vec can be implemented in DL4JTensorFlow
  
 +==To research==
 +*Implementation
 +*Cosine similarity, dot product equation usage
  
 +==Links==
 +*http://​deeplearning4j.org/​word2vec
  
mind/word2vec.1474842182.txt.gz · Last modified: 2016/09/25 16:23 by bayb2
Back to top
CC Attribution-Share Alike 4.0 International
chimeric.de = chi`s home Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0