nessneldivad / visual-word2vec

Word2Vec + Principal Component Analysis + Clustering for low-dimensional semantic representation of a set of words or compositional MWEs.

Word2Vec visualization

Word2Vec + Principal Component Analysis + Clustering for low-dimensional semantic representation of a set of words or compositional MWEs.

Requirements

Make sure you have at least 10GB of RAM available before running the script

Require python and the following packages :

gensim, numpy, scipy, matplotlib, sklearn, nltk (+ english stopwords dictionnary).

As well as the pre-trained word2vec model on Google News (heavy, decompress it in the same folder) : https://drive.google.com/file/d/0B7XkCwpI5KDYNlNUTTlSS21pQmM/edit

Usage

Enter words or MWEs > food,kitchen,delicious chicken,music,piano,saxophone,computer,screen,linux