Word2Vec is a natural language processing (NLP) technique developed by researchers at Google for learning word embeddings, which are dense vector representations of words in a continuous vector space. Word2Vec uses shallow neural networks to capture semantic and syntactic relationships between words based on their context in large text corpora.
It comes in two main architectures: Continuous Bag of Words (CBOW), which predicts a word based on its surrounding context, and Skip-Gram, which predicts the surrounding context given a word. These models enable Word2Vec to map similar words to nearby points in the vector space, allowing for operations such as analogical reasoning (e.g., “king” – “man” + “woman” ≈ “queen”). Word2Vec has significantly influenced the development of modern NLP by providing a foundation for more advanced models such as GloVe, FastText, and transformer-based architectures like BERT.
