Skip to content

Latest commit

 

History

History
10 lines (8 loc) · 477 Bytes

README.md

File metadata and controls

10 lines (8 loc) · 477 Bytes
Raw

Word2Vec implementation.

  • In keras, simple implementation.
  • In numpy, implemented with negative sampling. The OpinRank Hotel Review Dataset was used to create embeddings.

Results:

This image shows a plot of the 2 dimension PCA of the embeddings. These words were initially scattered randomly, and after training, they gradually formed clusters of similar words, as the image shows.

! Results Image