Word2vec is a model for learning vector representation of words called word embeddings. It transforms words in numerical form which then can be used in natural language processing and machine learning applications.
Word2Vec computes distributed vector representation of words. Distributed representations make the generalization to novel patterns easier and model estimation more robust. Distributed vector representation is used in natural language processing applications such as named entity recognition, disambiguation, parsing, tagging and machine translation.
Word2vec is a computationally effective predictive model for learning word embeddings from raw text. It uses unsupervised learning models, the Continuous Bag-of-Words model (CBOW) and the Skip-Gram model. Algorithmically, these models are the same, except that CBOW predicts target words from source context words, while the skip-gram does the inverse and predicts source context-words from the target words.
