#distributional-hypothesis

1 question from Text Representation & Embeddings

Timed test

1 question · 1 min · score and explanations at the end

Q 1 / 1Score 0 / 0

Shuffle questions

Question 1: Single choice

mediumAI-checkedText Representation & Embeddings

A word2vec model trained on news text finds that "hot" and "cold" are among each other's nearest neighbours. What explains this?

Antonyms appear in similar contexts, which is what word2vec measures1
The training corpus was too small for word2vec to tell the two words apart2
Negative sampling pushes words with opposite meanings towards each other3
The two words share most of their letters, so their vectors overlap4

Keys: 1–9 select · Enter check / next