WorksheetsML_Techniques_I Week 4 Session 1
Total questions: 5
Worksheet time: 3mins
Removing all stop-words from text always improves the performance of any downstream NLP model.
True
False
In CBOW (Continuous Bag of Words), the model predicts a context word given the center word.
True
False
In the Skip-Gram model, what is the primary training objective?
Predict the center word given its surrounding context
Predict surrounding context words given the center word
Maximize TF-IDF scores of rare words
Factorize the co-occurrence matrix
TF-IDF down-weights terms that:
Appear in many documents
Appear rarely across the corpus
Occur only in the test set
Are longer than a threshold
During an analogy test (“king” – “man” + “woman” ≈ “queen”), Word2Vec actually searches for
Euclidean distance to the target vector
Word frequency score
Cosine similarity to the target vector
Dot product with the centroid of all vector
