Continuous Bag of Words (CBOW) is a type of word embedding model introduced as part of the Word2Vec framework for learning distributed representations of words.
In CBOW, the model tries to predict the target word given its surrounding context words. Specifically, the context words are represented as one-hot vectors, and the goal is to learn an embedding vector for each word that maximizes the probability of predicting the central word.
This method captures semantic similarities between words, as words appearing in similar contexts tend to have similar embeddings. CBOW is computationally efficient and suitable for large datasets. Unlike the traditional Bag-of-Words, it retains some contextual information, making it useful in various downstream NLP tasks.
The CBOW model is widely used for generating dense, low-dimensional word representations.