Training Word2Vec with Gensim in NLP - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is Word2Vec?

Word2Vec is a method to turn words into numbers (vectors) so that computers can understand the meaning of words based on their context in sentences.

beginner

What are the two main architectures of Word2Vec?

The two main architectures are CBOW (Continuous Bag of Words) which predicts a word from its neighbors, and Skip-gram which predicts neighbors from a word.

intermediate

In Gensim, how do you start training a Word2Vec model on a list of sentences?

You create a Word2Vec object with parameters like vector_size and window, then call the .build_vocab() method with your sentences, and finally call .train() to train the model.

beginner

What does the 'window' parameter control in Word2Vec training?

The 'window' parameter controls how many words before and after the target word are considered as context during training.

intermediate

How can you check the similarity between two words using a trained Word2Vec model in Gensim?

Use the model's .wv.similarity('word1', 'word2') method to get a score showing how similar the two words are based on their vectors.

Which Gensim method is used to prepare the vocabulary before training Word2Vec?

A fit()

B train()

C build_vocab()

D init_vocab()

The build_vocab() method scans the sentences to build the vocabulary needed for training.

What does the 'vector_size' parameter specify in Word2Vec?

A Number of words in the vocabulary

B Size of the training batch

C Number of training epochs

D Length of the word vectors

vector_size sets how many numbers each word vector will have, defining its length.

Which Word2Vec architecture predicts the center word from surrounding words?

A CBOW

B RNN

C Skip-gram

D Transformer

CBOW (Continuous Bag of Words) predicts the center word from its context words.

How do you save a trained Word2Vec model in Gensim?

A model.export('filename')

B model.save('filename')

C model.write('filename')

D model.store('filename')

The save() method stores the model to disk for later use.

What type of data does Word2Vec expect for training?

A List of sentences, each sentence is a list of words

B Single long string of text

C Dictionary of word counts

D List of word vectors

Word2Vec trains on a list of sentences, where each sentence is a list of words (tokens).

Explain how to train a Word2Vec model using Gensim starting from raw text data.

Think about the steps from raw text to a trained model.

Describe the difference between CBOW and Skip-gram architectures in Word2Vec.

Focus on what each architecture tries to predict.