NLP and the Four Main Categories of Ambiguity DATA-SCIENCE-OPTIMIZE Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   What causes syntactic ambiguity in a sentence?

   • AThe sentence relies on the listener's knowledge of the speaker's intent
   • BA word in the sentence has multiple meanings
   • CThe words in the sentence are combined in an unusual way
   • DThe sentence can be parsed into more than one grammatical structure
   Reveal Solution Discussion Select an answer first
2. 2expert · hard

   A sentiment-analysis system for financial news must interpret the sentence: 'The stock market had a sharp decline.' The system must decide whether 'sharp' means 'sudden' or 'severe.' Which type of ambiguity is this, and what is the best way to resolve it?

   • ASemantic ambiguity; resolve by analyzing the combination of 'sharp' and 'decline'
   • BPragmatic ambiguity; resolve by considering the speaker's intent
   • CSyntactic ambiguity; resolve by parsing the sentence structure
   • DLexical ambiguity; resolve by using a dictionary of word senses
   Reveal Solution Discussion Select an answer first
4. 3expert · hard

   A voice assistant must respond to the user command: 'Turn on the light.' The assistant must decide whether the user is referring to a specific light in the room or the main light. Which type of ambiguity is this, and what information is most useful for resolving it?

   • ALexical ambiguity; the word 'light' has multiple meanings
   • BSyntactic ambiguity; the sentence can be parsed in multiple ways
   • CPragmatic ambiguity; the intended referent depends on the context and user's environment
   • DSemantic ambiguity; the meaning of 'turn on' is unclear
   Reveal Solution Discussion Select an answer first
5. 4expert · hard

   A customer-service chatbot is being designed to handle ambiguous user inputs. The team has a limited budget and must choose between two approaches: (1) implementing a word-sense disambiguation module that uses a dictionary of word senses, or (2) training a transformer-based model on a large corpus of customer-service conversations. The chatbot must handle both lexical ambiguity (e.g., 'bill' as a statement or a legislative proposal) and pragmatic ambiguity (e.g., 'Can you help me?' as a question or a request). Which approach is more likely to address both types effectively?

   • ANeither approach can handle pragmatic ambiguity, so the chatbot should be rule-based
   • BBoth approaches are equally effective, so the choice depends on cost
   • CThe transformer-based model, because it learns contextual representations that capture both word senses and speaker intent
   • DThe dictionary-based WSD module, because it directly resolves word senses
   Reveal Solution Discussion Select an answer first
6. 5foundation · easy

   Why does ambiguity pose a significant challenge for NLP systems?

   • AIt prevents NLP systems from processing any natural language at all
   • BIt can lead to incorrect interpretations of text, affecting tasks like translation and sentiment analysis
   • CIt makes text data impossible to store in databases
   • DIt only affects speech recognition, not text-based NLP
   Reveal Solution Discussion Select an answer first