Constituency Parsing Quiz

Answers and explanations appear as you take the quiz.

1. What does constituency parsing primarily produce for a sentence?
   • A tree of nested phrases like noun phrases and verb phrases
   • A list of word frequencies
   • A set of synonyms for each word
   • A translation into another language
2. How does constituency parsing differ from dependency parsing?
   • It only works on English
   • It groups words into phrases rather than linking words directly to each other
   • It ignores grammar entirely
   • It counts characters instead of words
3. Which classic algorithm is associated with constituency parsing using context-free grammars?
   • Dijkstra's algorithm
   • Gradient descent
   • The CYK (CKY) algorithm
   • Quicksort
4. What is at the root of a typical constituency parse tree?
   • A single word
   • A punctuation mark
   • A verb only
   • The S (sentence) node
5. Which dataset is famously used to train and evaluate English constituency parsers?
   • The Penn Treebank
   • ImageNet
   • MNIST
   • Common Crawl images
6. In the sentence 'The cat sat on the mat', what type of phrase is 'The cat'?
   • A verb phrase (VP)
   • A noun phrase (NP)
   • A prepositional phrase (PP)
   • An adjective phrase
7. What does an internal node in a constituency parse tree represent?
   • A word’s audio timestamp
   • A link between two unrelated documents
   • A labeled phrase spanning a contiguous part of the sentence
   • A predicted translation of the sentence
8. What does a span-based neural constituency parser score?
   • Only the sentence’s punctuation frequency
   • The speaker identity for each word
   • The similarity between two audio waveforms
   • Candidate spans and their possible phrase labels