Xplore Word Embedding Using CBOW Model and Skip-Gram Model

Word embedding has gained a lot of attention in recent years. To learn the systematic relation between words from a large junk of unlabeled text data is quite popular these days while using in Machine Translation, Question Answering or even in Text summarization. Computer understand only 0’s and 1’s now can understand text by simply using word embedding and making vectors in corpus i.e. word vector to represent them in numerical form which are used to perform linear algebra on the words which are now in numerical form in vectors. Word embedding is very effective tool in natural language processing (NLP), both CBOW (Continuous bag of words) and SKIP GRAM have increased the efficiency of word embedding along with it both models have reduced the training time. This paper is divided into two parts. In the first part we have discussed what is word embedding along with two models to implement word embedding, these models are continuous bag of word and skip gram. Our objective is to implement these two-model using shallow neural networks. In second part we have compared the efficiency of both the model while using different size of data sets.