Which approach uses a center word to predict surrounding context words?

Which approach uses a center word to predict surrounding context words?

• A. Continuous Bag of Words
• B. Continuous Skip-Gram
• C. Word2Vec CBOW
• D. Latent Dirichlet Allocation

Answer: (B)

Centering on one word to predict its surrounding context words is the Continuous Skip-Gram approach. In this setup, for each word in a text, you use that word as input and try to predict each nearby word within a defined window. The model’s objective is to maximize the likelihood of the actual context words given the center word, so the learned word vectors capture semantic and syntactic relationships based on those co-occurrences. This differs from Continuous Bag of Words, where the surrounding words are used to predict the center word, and from Latent Dirichlet Allocation, which is a topic-modeling method rather than a word-prediction model.

Centering on one word to predict its surrounding context words is the Continuous Skip-Gram approach. In this setup, for each word in a text, you use that word as input and try to predict each nearby word within a defined window. The model’s objective is to maximize the likelihood of the actual context words given the center word, so the learned word vectors capture semantic and syntactic relationships based on those co-occurrences. This differs from Continuous Bag of Words, where the surrounding words are used to predict the center word, and from Latent Dirichlet Allocation, which is a topic-modeling method rather than a word-prediction model.