ML B2 CH2

What is the main goal of Natural Language Processing (NLP)?

• To translate only programming languages
• To allow humans to code faster
• To help computers understand human language
• To improve grammar checking

Answer explanation

NLP focuses on enabling computers to understand and process human (natural) language.

Which of the following best describes a 'thesaurus' in NLP?

• A grammar checker
• A list of antonyms
• A collection of synonymous words
• A programming dictionary

Answer explanation

A thesaurus groups words with similar or identical meanings, which helps in NLP tasks.

What is a major disadvantage of using a thesaurus for NLP?

• It can understand subtle word differences
• It evolves with language changes
• It relies heavily on manual human effort
• It uses complex algorithms

Answer explanation

Thesauri are labor-intensive and struggle to keep up with evolving language.

What is a corpus in NLP?

• A spell checker
• A dataset of program codes
• A large set of structured programming rules
• A large collection of text used to train NLP models

Answer explanation

A corpus is a dataset of natural language text used for analysis and model training.

According to the distributional hypothesis, the meaning of a word is:

• Defined by a dictionary
• Based on spelling rules
• Formed by the words around it
• Determined by its length

Answer explanation

The distributional hypothesis states that a word’s meaning is derived from its context.

What is a co-occurrence matrix?

• A matrix that shows the frequency of sentences
• A table of synonym relationships
• A way to calculate sentence structure
• A table counting surrounding words of a target word

Answer explanation

Co-occurrence matrices count how often words appear near each other in text.

Which method is used to measure similarity between word vectors?

• Euclidean Distance
• Cosine Similarity
• Manhattan Distance
• Hamming Distance

Answer explanation

Cosine similarity measures the angle between vectors, indicating similarity in direction.

What does a Cosine Similarity value of 1 mean?

• The words are opposite in meaning
• The words are completely unrelated
• The vectors are in the same direction (high similarity)
• The vectors are perpendicular

Answer explanation

A value of 1 in cosine similarity means the word vectors point in the same direction.

What problem does Positive Pointwise Mutual Information (PPMI) solve?

• It increases the word frequency
• It reduces vector dimensionality
• It avoids negative values in co-occurrence statistics
• It creates a better dictionary

Answer explanation

PPMI prevents negative values that could arise in raw PMI calculations.

What does Singular Value Decomposition (SVD) help with in NLP?

• Making sentences longer
• Increasing vocabulary size
• Reducing the dimensionality of word vectors
• Building thesaurus databases

Answer explanation

SVD reduces large word vector spaces into smaller, meaningful representations.