word2vector icon indicating copy to clipboard operation
word2vector copied to clipboard

words in different paragraphs

Open mngyng opened this issue 6 years ago • 1 comments

Does this one take care of words in different paragraphs? If not, is there a workaround to do it?

mngyng avatar Jan 14 '19 01:01 mngyng

Hi @mngyng, Before using word2vector, you should preprocess your documents into words which are split by spaces. The key thing that word2vector cares is the "distance" between words. For more information please check other papers about NN(Neural network). Thanks :)
Here is some reference:

  1. Distributed Representations of Sentences and Documents
  2. Efficient Estimation of Word Representations in Vector Space
  3. word2vec Explained: Deriving Mikolov et al.’s

LeeXun avatar Jan 17 '19 10:01 LeeXun