learn.ifs.name

Semantic Role Labeling

Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 22

7 cards in “22.4 PropBank” · whole deck

22.4 PropBank

  1. What is PropBank, and how are its roles defined?
    The Proposition Bank: sentences (e.g. the Penn Treebank) annotated with semantic roles defined per verb sense and named only by number: Arg0 is generally the PROTO-AGENT, Arg1 the PROTO-PATIENT, higher numbers verb-specific.
  2. What do PropBank’s Arg2–Arg4 typically mean?
    Less consistently: Arg2 often benefactive, instrument, attribute or end state; Arg3 start point, benefactive, instrument or attribute; Arg4 the end point.
  3. What is a PropBank frame file? Give the entry for fall.
    A per-verb-sense list of roles with informal human-readable glosses. fall.01 “move downward”: Arg1 thing falling, Arg2 extent/amount fallen, Arg3 start point, Arg4 end point. “[Arg1 Sales] fell [Arg4 to $25 million] [Arg3 from $27 million].”
  4. Why does fall have no Arg0?
    Its normal subject is a patient-like PROTO-PATIENT, not an agent.
  5. What does PropBank labeling reveal about “Big Fruit Co. increased the price of bananas”, “The price of bananas was increased by Big Fruit Co.” and “The price of bananas increased 5%”?
    In each, Big Fruit Co. is Arg0 (agent) and the price of bananas Arg1 (theme), despite different syntax; 5% is Arg2, the amount increased.
  6. What are PropBank ArgMs?
    Non-numbered modifier/adjunct arguments stable across predicates: ArgM-TMP (time), ArgM-LOC (location), ArgM-DIR (direction), ArgM-MNR (manner), ArgM-PRP/CAU (purpose, cause), ArgM-NEG (negation), etc.
  7. What is NomBank?
    A PropBank-style annotation of noun predicates: in Apple’s agreement with IBM, Apple is Arg0 and IBM Arg2.

Source