{"id":16817,"date":"2018-06-05T05:40:09","date_gmt":"2018-06-05T05:40:09","guid":{"rendered":"https:\/\/data-flair.training\/blogs\/?p=16817"},"modified":"2021-05-14T11:00:14","modified_gmt":"2021-05-14T05:30:14","slug":"tensorflow-word2vec","status":"publish","type":"post","link":"https:\/\/data-flair.training\/blogs\/tensorflow-word2vec\/","title":{"rendered":"Word2Vec: TensorFlow Vector Representation Of Words"},"content":{"rendered":"

In this TensorFlow<\/strong> article “Word2Vec: TensorFlow Vector Representation Of Words”, we\u2019ll be looking at a convenient method of representing words as vectors, also known as word embeddings. <\/span><\/p>\n

Moreover, in this TensorFlow word embedding tutorial, we will be looking at scaling with noise-induced training & Word2Vec skip gram model, Along with this, we will discuss how to build graph and training for TensorFlow Word2Vec and also examples of Word2Vec in TensorFlow.<\/span><\/p>\n

So, let’s start Word2Vec in TensorFlow.<\/p>\n

Vector Representation of Words<\/span><\/h2>\n

Most of the models that you\u2019ll work with will have high-dimensional vectors of datasets ranging from pixel values in raw images to power densities in audio tracks. <\/span><\/p>\n

Now, all the useful information that your image<\/strong> and audio recognition models<\/strong> will need is in this raw data whereas, in the case of working with natural language processing, words are treated as symbols, a Word2Vec example being ‘cat’ that is taken as\u00a0<\/span>Id537<\/span>\u00a0and ‘dog’ as\u00a0<\/span>Id143<\/span>. <\/span><\/p>\n

This allotment is random and provides no information if the data is interrelated. Therefore, representing words as individual identities may lead to sparsity in the dataset, which increases the need for more data for your model. Vector representation of data can, therefore, let you tackle these problems.<\/span><\/p>\n

\"Word2vec-<\/a>

Word2vec- (Word Embedding)<\/p><\/div>\n

<\/div>\n
VSMs or Vector Space Models embed nearby these words in a vector space.\u00a0<\/span><\/div>\n
\n
We typically use VSMs with natural language processing but it is also possible to use them elsewhere as per the need.<\/div>\n<\/div>\n

The distribution hypothesis, tells that the choice of words in the same context share a common meaning and this can be used by dividing the approach into two categories:\u00a0<\/span><\/p>\n