74.419 Artificial Intelligence 2004

1
74.419 Artificial Intelligence 2004

• From Syntax to Semantics
• Grammatical Extensions
• Sentence Structures
• Noun Phrase - Modifications
• Verb Phrase - Subcategorization
• Feature Structures
• ?-expressions

2
Sample Grammar
Grammar (S, NT, T, P) Part-of-Speech ? NT,
syntactic Constituents ? NT S ? NP
VP statement S ? Aux NP VP question S ?
VP command NP ? Det Nominal NP ? Proper-Noun
Nominal ? Noun Noun Nominal Nominal PP VP ?
Verb Verb NP Verb PP Verb NP PP PP ? Prep
NP Det ? that this a Noun ? book flight
meal money Proper-Noun ? Houston American
Airlines TWA Verb ? book include prefer Aux
? does Prep ? from to on Task Parse "Does
this flight include a meal?"
3
Sample Parse Tree
Task Parse "Does this flight include a
meal?" S Aux NP
VP Det Nominal Verb NP
Noun Det Nominal does this
flight include a meal
4
Grammar Sentence Level Constructs

• Sentence Level Constructs
• declarative S ? NP VP
• This flight leaves at 9 am.
• imperative S ? VP
• Book this flight for me.
• yes-no-question S ? Aux NP VP
• Does this flight leave at 9 am?
• wh-question S ? Wh-NP Aux NP VP
• When does this flight leave Winnipeg?

5
Grammar Noun Phrase Modification 1

• Noun Phrase Modifiers
• head the central noun of the NP ( modifiers)
• modifiers before the head noun (prenominal)
• determiner the, a, this, some, ...
• predeterminer all the flights
• cardinal numbers, ordinal numbers one flight, the
  first flight, ...
• quantifiers much, little
• adjectives a first-class flight, a long flight
• adjective phrase the least expensive flight
• NP ? (Det) (Card) (Ord) (Quant) (AP) Nominal

6
Grammar Noun Phrase Modification 2

• Noun Phrase Modifiers (continued)
• modifiers after the head noun (post-nominal)
• prepositional phrase PP all flights from Chicago
• Nominal ? Nominal PP (PP) (PP)
• non-finite clause, gerundive postmodifers all
  flights arriving after 7 pm
• Nominal ? GerundVP
• GerundVP ? GerundV NP GerundV PP ...
• relative clause a flight that serves breakfast
• Nominal ? Nominal RelClause
• RelClause ? (who that) VP

7
Grammar Verb Subcategorization

• Verb Phrase and Subategorization
• VP Verb other constituents. Different verbs
  accept or need different constituents ? Verb
  Subcategorization captured in verb frames.
• sentential complement VP ? Verb inf-sentence
• I want to fly from Boston to Chicago.
• NP complement VP ? Verb NP
• I want this flight.
• no complement VP ? Verb
• I sleep.
• more forms VP ? Verb PP PP
• I fly from Boston to Chicago.

8
Grammar Feature Structures 1

• Feature Structures
• describe additional syntactic-semantic
  information, like category, person, number,
  e.g. goes ? (verb, 3rd, singular)
• specify feature structure constraints
  (agreements) as part of grammar
• during parsing, check agreements of feature
  structures (unification)
• example
• S ? NP VP ltNP numbergt ltVP numbergt
• or S ? NP VP ltNP agreementgt ltVP agreementgt

9
Grammar Feature Structures 2

• Feature Structures
• agreements in general determined by head of
  phrase, i.e. central noun or verb
• example
• (1) ... the man who chased the cat out of
  the house ...
• central noun?
• (2) ... the man chased the barking dog who
  bit him ...
• central verb?
• operations on agreements (e.g. sing _ plural
  fail)
• unification for checking of specified agreements

10
Semantics

• Distinguish between surface structure (syntactic
  structure) and deep structure (semantic
  structure) of sentences.
• Semantics can be defined based on FOPL.
• Transform sentence (with quantified variables)
  into lambda-expression. Central is again verb
  (with roles).
• lambda-expression is like a function which can
  then be applied to constants.
• example ?x, y loves (x, y) FOPL sentence
• ?-expr ?x?y loves (x, y) function
• ?x?y loves (x, y) (John) ? ?y loves (John,
  y)
• Note The semantics of LISP is based on
  lambda-calculus.

11
Semantics

• Distinguish between surface structure (closer to
  syntactic structure) and deep structure (semantic
  structure) of sentences
• Semantic can be defined based on FOPL.
• Transform sentence (with quantified variables)
  into lambda-expression. Central is again verb
  (with roles).
• lamda-expression is like a function which can
  then be applied to constants.
• AI Caramba is close to ICSI.
• specific close-to (AI Caramba, ICSI)
• general ?x,y close-to (x, y) ? xAI Caramba ?
  yICSI
• ?-expr ?x?y close-to (x, y) (AI Caramba)
• ?y close-to (AI Caramba, y)
• close-to (AI Caramba, ICSI)

12
Additional References

• Jurafsky, D. J. H. Martin, Speech and Language
  Processing, Prentice-Hall, 2000. (Chapters 9 and
  10)

Earley Algorithm Jurafsky Martin, Figure
10.16, p.384
Earley Algorithm - Examples Jurafsky Martin,
Figures 10.17 and 10.18