Bird
0
0

You want to compare t-SNE with PCA for embedding visualization. Which statement is true?

hard📝 Application Q9 of 15
NLP - Word Embeddings
You want to compare t-SNE with PCA for embedding visualization. Which statement is true?
At-SNE is faster than PCA on large datasets
BPCA is nonlinear, t-SNE is linear
Ct-SNE captures local structure better, PCA preserves global variance
DPCA requires labeled data, t-SNE does not
Step-by-Step Solution
Solution:

  1. Step 1: Recall PCA and t-SNE characteristics

    PCA is a linear method preserving global variance; t-SNE is nonlinear and preserves local neighborhoods.

  2. Step 2: Evaluate each option

    Only t-SNE captures local structure better, PCA preserves global variance correctly describes their differences.

  3. Final Answer:

    t-SNE captures local structure better, PCA preserves global variance -> Option C

  4. Quick Check:

    t-SNE local, PCA global structure [OK]
Quick Trick: t-SNE = local, PCA = global structure [OK]
Common Mistakes:
MISTAKES
  • Mixing linearity of methods
  • Assuming t-SNE is faster
  • Thinking PCA needs labels

Want More Practice?

15+ quiz questions · All difficulty levels · Free

Free Signup - Practice All Questions
More NLP Quizzes
Sentiment Analysis Advanced - Fine-grained sentiment (5-class) - Quiz 7medium Sentiment Analysis Advanced - Sentiment with context (sarcasm, negation) - Quiz 6medium Sequence Models for NLP - Embedding layer usage - Quiz 9hard Sequence Models for NLP - Why sequence models understand word order - Quiz 1easy Sequence Models for NLP - RNN for text classification - Quiz 9hard Text Similarity and Search - Jaccard similarity - Quiz 4medium Topic Modeling - Why topic modeling discovers themes - Quiz 14medium Word Embeddings - Word similarity and analogies - Quiz 8hard Word Embeddings - Pre-trained embedding usage - Quiz 15hard Word Embeddings - Pre-trained embedding usage - Quiz 9hard