Sunteți pe pagina 1din 54

 AI agents deal with knowledge (data)

◦ Facts (believe & observe knowledge)


◦ Procedures (how to knowledge)
◦ Meaning (relate & define knowledge)
 Right representation is crucial
◦ Early realisation in AI
◦ Wrong choice can lead to project failure
◦ Active research area.

Biomedical Dept, NIT Raipur 10/11/09 1


 Requirements:
◦ Adequacy (I) (also called completeness)
◦ Correctness (II)
◦ Efficiency (III)
I/II/III

Representational Inferential Acquisitional


(learning)

Biomedical Dept, NIT Raipur 10/11/09 2


 Logic is concerned with reasoning & validity
of arguments,
 Logic may be or may not be true but should

be valid.
 For example: All lemons are blue.

Mary is Lemon
Therefore Marry is blue.
Logic is used as representational method for
AI.

Biomedical Dept, NIT Raipur 10/11/09 3


 In this valid statements are determined
according to rules of propositional syntax.

 Syntax governs combination of basic


building blocks such as propositions &
logical connectives.

 Propositions are elementary atomic


sentences.

Biomedical Dept, NIT Raipur 10/11/09 4


It is raining.
Snow is white.
My car is painted silver. Etc.

Compound propositions are formed from atomic formulas


using logical connectives such as
not,
and,
or,
if,
iff.
It is raining & the wind is blowing.
If you study hard you will be rewarded.
The sum of 20 & 30 is not 60

Biomedical Dept, NIT Raipur 10/11/09 5


 - for not or negation.
 & or ⋀ for and or conjunction.
 V for or or disjunction.
 → for if …….. then implication.
 ↔ for if and only if or double implication.

Example: 1. It is raining and wind is blowing.


Syntax : (R&B)
2. It is raining and it is Tuesday.
R ⋀T

Biomedical Dept, NIT Raipur 10/11/09 6


 Syntax
◦ Rules for constructing legal sentences in the logic
◦ Which symbols we can use (English: letters, punctuation)
◦ How we are allowed to combine symbols
 Semantics:The semantics or meaning of a sentence is
just a value True or False
◦ How we interpret (read) sentences in the logic
◦ Assigns a meaning to each sentence (True or false)
 Example: “All lecturers are seven foot tall”
◦ A valid sentence (syntax): (L→ 7)
◦ And we can understand the meaning (semantics)( False)(F)
◦ This sentence happens to be false (there is a counterexample)

Biomedical Dept, NIT Raipur 10/11/09 7


 It is raining RAINING
 It is sunny SUNNY
 It is windy WINDY
 If it is rainy then it is not sunny

RAINY→ - SUNNY
 It is not raining in NEW YORK

-R(N)
 I am either not well or just very tired

-W(I) ⋁ T(I)

Biomedical Dept, NIT Raipur 10/11/09 8


1. Whenever he eats sandwiches that have
pickles in them, he ends up either asleep
at his desk or singing loud songs.
S(y)⋀E(x,y) ⋀ P(y) →A(x) ⋁(S(x,z) ⋀ L(z))
S(y) = y is a sandwich.
E(x,y) = x(man) eats y( sandwich)
P(y) = y(sandwich) has pickles in it
A(x) = x ends with sleep
S(x,z) = x(man) sings z(songs)
L(z) = means that z(the songs) are Loud.

Biomedical Dept, NIT Raipur 10/11/09 9


 First-order logic is used to model the world in terms of
◦ objects which are things with individual identities
e.g., individual students, lecturers, companies,
cars ...
◦ properties of objects that distinguish them from
other objects
e.g., mortal, blue, oval, even, large, ...
◦ classes of objects (often defined by properties)
e.g., human, mammal, machine, ...
◦ relations that hold among objects
e.g., brother of, bigger than, outside, part of, has
color, occurs after, owns, a member of, ...
◦ functions which are a subset of the relations in
which there is only one ``value'' for any given
``input''.
e.g., father of, best friend, second half, one more
than ...
Biomedical Dept, NIT Raipur 10/11/09 10
 Predicates: P(x[1], ..., x[n])
◦ P: predicate name; (x[1], ..., x[n]): argument
list
◦ A special function with range = {T, F};
◦ Examples: human(x), /* x is a human */
father(x, y) /* x is the father of y */
◦ When all arguments of a predicate is assigned values
(said to be instantiated), the predicate becomes either
true or false, i.e., it becomes a proposition. Ex.
Father(Fred, Joe)

 Terms (arguments of predicates must be terms)


◦ Constants are terms (e.g., Fred, a, Z, “red”, etc.)
◦ Variables are terms (e.g., x, y, z, etc.), a variable is
instantiated when it is assigned a constant as its value
◦ Functions of terms are terms (e.g., f(x, y, z), f(x, g(a)),
etc.)
◦ A term is called a ground term if it does not involve
variables

Biomedical Dept, NIT Raipur 10/11/09 11


 Quantifiers
Universal quantification ∀ (or forall)
◦ (∀x)P(x) means that P holds for all values of x in the
domain associated with that variable.
◦ E.g., (∀x) dolphin(x) => mammal(x)
(∀x) human(x) => mortal(x)
◦ Universal quantifiers often used with "implication (=>)"
to form "rules" about properties of a class
(∀x) student(x) => smart(x) (All students
are smart)
◦ Often associated with English words “all”, “everyone”,
“always”, etc.
◦ You rarely use universal quantification to make blanket
statements about every individual in the world
(because such statement is hardly true)
(∀x)student(x)^smart(x) means everyone
in the world is a student and is smart.

Biomedical Dept, NIT Raipur 10/11/09 12


Existential quantification ∃
◦ (∃ x)P(x) means that P holds for some value(s) of x in
the domain associated with that variable.
◦ E.g., (∃ x) mammal(x) ^ lays-eggs(x)
(∃ x) taller(x, Fred)
(∃ x) UMBC-Student (x) ^ taller(x, Fred)
◦ Existential quantifiers usually used with “^ (and)" to
specify a list of properties about an individual.
(∃ x) student(x) ^ smart(x) (there is a student who is
smart.)
◦ A common mistake is to represent this English
sentence as the FOL sentence:
(∃ x) student(x) => smart(x)
It also holds if there no student exists in the domain
because
student(x) => smart(x) holds for any individual who is
not a student.
◦ Often associated with English words “someone”,
◦ “sometimes”, etc.

Biomedical Dept, NIT Raipur 10/11/09 13


 Each quantified variable has its scope
◦ (∀x)[human(x) => (∃ y) [human(y) ^ father(y, x)]
◦ All occurrences of x within the scope of the
quantified x refer to the same thing.
◦ Use different variables for different things
 Switching the order of universal quantifiers does
not change the meaning:
◦ (∀x)(∀y)P(x,y) <=> (∀y)(∀x)P(x,y), can write as
(∀x,y)P(x,y)
 Similarly, you can switch the order of existential
quantifiers.
◦ (∃ x)(∃ y)P(x,y) <=> (∃ y)(∃ x)P(x,y)
 Switching the order of universals and existential
does change meaning:
◦ Everyone likes someone: (∀x)(∃ y)likes(x,y)
◦ Someone is liked by everyone: (∃ y)(∀x) likes(x,y)
Biomedical Dept, NIT Raipur 10/11/09 14
 Every gardener likes the sun.
(∀x) gardener(x) => likes(x,Sun)
 Not Every gardener likes the sun.
~((∀x) gardener(x) => likes(x,Sun))
 You can fool some of the people all of the time.
(∃ x)(∀t) person(x) ^ time(t) => can-be-fooled(x,t)
 You can fool all of the people some of the time.
(∀x)(∃ t) person(x) ^ time(t) => can-be-fooled(x,t)
(the time people are fooled may be different)
 You can fool all of the people at some time.
(∃ t)(∀x) person(x) ^ time(t) => can-be-fooled(x,t)
(all people are fooled at the same time)
 You can not fool all of the people all of the time.
~((∀x)(∀t) person(x) ^ time(t) => can-be-fooled(x,t))
 Everyone is younger than his father
 (∀x) person(x) => younger(x, father(x))

Biomedical Dept, NIT Raipur 10/11/09 15


 All purple mushrooms are poisonous.
(∀x) (mushroom(x) ^ purple(x)) =>
poisonous(x)
 No purple mushroom is poisonous.
~(∃ x) purple(x) ^ mushroom(x) ^ poisonous(x)
(∀x) (mushroom(x) ^ purple(x)) =>
~poisonous(x)
 There are exactly two purple mushrooms.
(∃ x)(Ey) mushroom(x) ^ purple(x) ^
mushroom(y) ^ purple(y) ^ ~(x=y) ^
(∀z) (mushroom(z) ^ purple(z)) => ((x=z) v
(y=z))
 Clinton is not tall.
~tall(Clinton)
 X is above Y if X is directly on top of Y or
there is a pile of one or more other
objects directly on top of one another
starting with X and ending with Y.
(∀x)(∀y) above(x,y) <=> (on(x,y) v (∃ z)
(on(x,z) ^ above(z,y)))
Biomedical Dept, NIT Raipur 10/11/09 16
 Build a small genealogy knowledge base by
FOL that
◦ contains facts of immediate family relations
(spouses, parents, etc.)
◦ contains definitions of more complex relations
(ancestors, relatives)
◦ is able to answer queries about relationships
between people
 Predicates:
◦ parent(x, y), child (x, y), father(x, y), daughter(x,
y), etc.
◦ spouse(x, y), husband(x, y), wife(x,y)
◦ ancestor(x, y), descendent(x, y)
◦ relative(x, y)
 Facts:
◦ husband(Joe, Mary), son(Fred, Joe)
◦ spouse(John, Nancy), male(John), son(Mark, Nancy)
◦ father(Jack, Nancy), daughter(Linda, Jack)
◦ daughter(Liz, Linda)
◦ etc.
Biomedical Dept, NIT Raipur 10/11/09 17
 Rules for genealogical relations
◦ (∀x,y) parent(x, y) <=> child (y, x)
(∀x,y) father(x, y) <=> parent(x, y) ^ male(x) (similarly for
mother(x, y))
(∀x,y) daughter(x, y) <=> child(x, y) ^ female(x) (similarly for
son(x, y))
◦ (∀x,y) husband(x, y) <=> spouse(x, y) ^ male(x) (similarly for
wife(x, y))
(∀x,y) spouse(x, y) <=> spouse(y, x) (spouse relation is
symmetric)
◦ (∀x,y) parent(x, y) => ancestor(x, y)
(∀x,y)(∃ z) parent(x, z) ^ ancestor(z, y) => ancestor(x, y)
◦ (∀x,y) descendent(x, y) <=> ancestor(y, x)
◦ (∀x,y)(∃ z) ancestor(z, x) ^ ancestor(z, y) => relative(x, y)
(related by common ancestry)
(∀x,y) spouse(x, y) => relative(x, y) (related by marriage)
(∀x,y)(∃ z) relative(z, x) ^ relative(z, y) => relative(x, y)
(transitive)
(∀x,y) relative(x, y) => relative(y, x) (symmetric)
 Queries
◦ ancestor(Jack, Fred) /* the answer is yes */
◦ relative(Liz, Joe) /* the answer is yes */
◦ relative(Nancy, Mathews)
/* no answer in general, no if under closed world
assumption */

Biomedical Dept, NIT Raipur 10/11/09 18


 “It is not the case that everyone is ...” is
logically equivalent to “There is someone
who is NOT ...”
 “No one is ...” is logically equivalent to “All
people are NOT ...”
 We can relate sentences involving forall
and exists using De Morgan’s laws:
~(∀x)P(x) <=> (∃ x) ~P(x)
~(∃ x) P(x) <=> (∀x) ~P(x)
(∃ x) P(x) <=> ~(∀x) ~P(x)
(∀x) P(x) <=> ~ (∃ x) ~P(x)
 Example: no one likes everyone
◦ ~ (∃ x)(∀y)likes(x,y)
◦ (∀x)(∃ y)~likes(x,y)
Biomedical Dept, NIT Raipur 10/11/09 19
Biomedical Dept, NIT Raipur 10/11/09 20
 Clause is disjunction of number of literals.
 A Ground clause is one in which no

variables occur in the expression.


 A Horn clause is a clause with at most one

positive literal.

Biomedical Dept, NIT Raipur 10/11/09 21


 A term (denoting a individual in the world) is a constant
symbol, a variable symbol, or a function of terms.
 An atom (atomic sentence) is a predicate P(x[1], ..., x[n])
◦ Ground atom: all terms in its arguments are ground terms
(does not involve variables)
◦ A ground atom has value true or false (like a proposition in
PL)
 A literal is either an atom or a negation of an atom
 A sentence is an atom, or,
◦ ~P, P v Q, P ^ Q, P => Q, P <=> Q, (P) where P and Q are
sentences
◦ If P is a sentence and x is a variable, then (∀x)P and (∃ x)P
are sentences
 A well-formed formula (wff) is a sentence containing no
"free" variables. i.e., all variables are "bound" by universal or
existential quantifiers.
(∀x)P(x,y) has x bound as a universally quantified variable,
but y is free.
Biomedical Dept, NIT Raipur 10/11/09 22
1. Eliminate →, using: a → b= ¬ a v b.
2. Reduce the scope of each ¬ to a single term, using:
¬ (¬ p) = p
deMorgan's laws: ¬(a Λ b) = ¬ a V ¬ b
¬(a V b) = ¬ a Λ ¬ b
¬ ∀x P(x) = ∃ x ¬ P(x)
¬ ∃ x P(x) = ∀x ¬ P(x)
1. Standardize variables.
2. Move all quantifiers to the left of the formula without
changing their relative order.
3. Eliminate existential quantifiers by inserting Skolem
functions.
4. Drop the prefix.
5. Convert the expression into a conjunction of disjuncts,
using associativity and distributivity.
6. Create a separate clause for each conjunct.
7. Standardize apart the variables in the set of clauses
generated in step 8, using the fact that: (∀x: P(x) Λ
Q(x)) = ∀x: P(x) Λ ∀x: Q(x)

Biomedical Dept, NIT Raipur 10/11/09 23


 Objective
◦ Want all variables universally quantified
◦ Notational variant of FOL w/o existentials
◦ Retain implicitly full FOL expressiveness
 Skolem’s Theorem
Every existentially quantified variable can be replaced
by a unique Skolem function whose arguments are all
the universally quantified variables on which the
existential depends, without changing FOL.
 Examples
◦ “Everybody likes something”
∀(x) ∃ (y) [Person(x) & Likes(x,y)]
∀(x) [Person(x) & Likes(x, S1(x))]
Where S1(x) = “that which x likes”
◦ “Every philosopher writes at least one book”
∀(x) ∃ (y)[Philosopher(x) & Book(y)) => Write(x,y)]
∀(x)[(Philosopher(x) & Book(S2(x))) => Write(x,S2(x))]

Biomedical Dept, NIT Raipur 10/11/09 24


 Example:
∀x: [Roman(x) Λ know(x, Marcus)] →[hate(x,Caesar) V
(∀y: ∃ z: hate(y,z) → thinkcrazy(x,y))]
◦ Eliminate →
∀x: ¬[Roman(x) Λ know(x, Marcus)] V
[hate(x,Caesar) V (∀y: ∃ z:¬( hate(y,z) V
thinkcrazy(x,y))]
◦ Reduce scope of ¬.
∀x: [ ¬Roman(x) V ¬ know(x, Marcus)] V [hate(x,Caesar) V
(∀y: ∀z: ¬hate(y,z) V thinkcrazy(x,y))]
◦ “Standardize” variables: Rename Variables:
∀x: P(x) V ∀x: Q(x) converts to ∀x: P(x) V ∀y: Q(y)
◦ Move quantifiers. ∀x: ∀y: ∀z: [¬Roman(x) V ¬ know(x,
Marcus)] V [hate(x,Caesar) V (¬hate(y,z) V thinkcrazy(x,y))]
Biomedical Dept, NIT Raipur 10/11/09 25
◦ Eliminate existential quantifiers.
∃ y: President(y) will be converted to
President(S1)
∀x: ∃ y: father-of(y,x) will be converted to ∀x:
father-of(S2(x),x))
◦ Drop the prefix.
[ ¬Roman(x) ¬ know(x,Marcus)] V [hate(x,
Caesar) V (¬ hate(y,z) V thinkcrazy(x,y))]
◦ Convert to a conjunction of disjuncts.
¬ Roman(x) V ¬ know(x,Marcus) V hate(x,Caesar) V ¬
hate(y,z) V thinkcrazy(x,y)

Biomedical Dept, NIT Raipur 10/11/09 26


 It is a simple iterative process : at each
step, two clauses, called the parent clauses
are compared(resolved), yielding a new
clause that has been inferred from them.

 The new clause represents the ways that


two parent clauses interact with each other.

Biomedical Dept, NIT Raipur 10/11/09 27


 Given:
winter V summer
¬ winter V cold
We can conclude:
summer v cold
 Herbrand's Theorem:
To show that a set of clauses S is unsatisfiable,
it is necessary to consider only interpretations
over a particular set, called the Herbrand
universe of S. A set of clauses S is unsatisfiable
if and only if a finite subset of ground instances
(in which all bound variables have had a value
substituted for them) of S is unsatsifable.

Biomedical Dept, NIT Raipur 10/11/09 28


1. Convert all the propositions of F to clause form.
2. Negate P and convert the result to clause form.
Add it to the set of clauses obtained in step 1.
3. Repeat until either a contradiction is found or no
progress can be made:
a) Select two clauses. Call these the parent clauses.
b) Resolve them together. The resolvent will be the
disjunction of all of the literals of both of the parent
clauses with the following exception: If there are any
pairs of literals L and ¬ L such that one of the parent clauses
contains L and the other contains ¬L, then select one such pair
and eliminate both L and ¬L from the resolvent.
c) If the resolvent is the empty clause, then a contradiction has
been found. If it is not, then add it to the set of clauses
available to the procedure.

Biomedical Dept, NIT Raipur 10/11/09 29


 Consider the given Axioms Given in table of next
slides (First Column).
 Suppose we want to prove R.
 First we convert each axiom in to clause form.
 Negate R i.e ¬R.
 Select pair of clauses to resolve together.
 The result gives an empty clause. This is
called resolution proof by refutition.

Biomedical Dept, NIT Raipur 10/11/09 30


Given Axioms Clause Form
P P (1)

(P Λ Q) → R ¬P V ¬Q V R (2)

(S V T) → Q ¬S V Q (3)

¬T V Q (4)

T T (5)

Biomedical Dept, NIT Raipur 10/11/09 31


¬P V ¬Q V R ¬R

¬P V ¬Q P

¬T V Q ¬Q

¬T T

Biomedical Dept, NIT Raipur 10/11/09 32


 Axioms in clause form:
1. man(Marcus)
2. Pompeian(Marcus)
3. ¬ Pompeian(x1) v Roman(x1)
4. Ruler(Caesar)
5. ¬ Roman(x2) v loyalto(x2, Caesar) v hate(x2,
Caesar)
6. loyalto(x3, f1(x3))
7. ¬ man(x4) v ¬ ruler(y1) v ¬ tryassassinate(x4,
y1) v
loyalto (x4, y1)
8. tryassassinate(Marcus, Caesar)
Biomedical Dept, NIT Raipur 10/11/09 33
Prove: hate(Marcus, Caesar) ¬hate(Marcus, Caesar) 5

Marcus/x2
3 ¬Roman(Marcus) V loyalto(Marcus,Caesar)

Marcus/x1
¬Pompeian(Marcus) V loyalto(Marcus,Caesar) 2

7 loyalto(Marcus,Caesar)

Marcus/x4, Caesar/y1

1 ¬man(Marcus) V ¬ ruler(Caesar) V ¬ tryassassinate(Marcus, Caesar)

¬ ruler(Caesar) V ¬ tryassassinate(Marcus, Caesar) 4

¬ tryassassinate(Marcus, Caesar) 8

Biomedical Dept, NIT Raipur 10/11/09 34


Prove: loyalto(Marcus, Caesar) ¬loyalto(Marcus, Caesar) 5

Marcus/x2

3 ¬Roman(Marcus) V hate(Marcus,Caesar)
Marcus/x1
¬Pompeian(Marcus) V hate(Marcus,Caesar) 2

hate(Marcus,Caesar)
(a)
hate(Marcus,Caesar) 10

Marcus/x6, Caesar/y3
persecute(Caesar, Marcus) 9
Marcus/x5, Caesar/y2
hate(Marcus,Caesar)
:
: (b)
Biomedical Dept, NIT Raipur 10/11/09 35
 Axioms in clause form:
1. man(Marcus)
2. Pompeian(Marcus)
3. Born(Marcus, 40)
4. ¬ man(x1) V mortal(x1)
5. ¬ Pompeian(x2) V died(x2,79)
6. erupted(volcano, 79)
7. ¬ mortal(x3) V ¬ born(x3, t1) V ¬gt(t2—t1, 150) V
dead(x3, t2)
8. Now=2002
9. ¬ alive(x4, t3) V ¬dead (x4, t3)
10. ¬ dead(x5, t4) V alive (x5, t4)
11. ¬ died (x6, t5) V ¬ gt(x6, t5) V dead(x6, t6)

Prove: ¬alive(Marcus, now)

Biomedical Dept, NIT Raipur 10/11/09 36


 Requires full formal representation in FOL
(for conversion to clause form)
 Resolution defines a search space (which
clauses will be resolved against which
others define the operators in the space) 
search method required
 Worst case: resolution is exponential in the
number of clauses to resolve. Actual:
exponential in average resolvable set (=
branching factor)
 Can we define heuristics to guide search for
BestFS, or A* or B*? (Not in the general
case)

Biomedical Dept, NIT Raipur 10/11/09 37


 Advantages of using logic
◦ No need to make a difference between
knowledge representation and the inference
method
◦ Soundness (a false statement can not be
derived) and completeness (all true statements
can be derived)
◦ Has a logical make-up
 Disadvantage of using logic
◦ The derivation takes a lot of effort
◦ Difficult to use different layers of representation

Biomedical Dept, NIT Raipur 10/11/09 38


 Logic programming
◦ New programming paradigm
◦ Viewing the set of clauses K as a program
 PROLOG
◦ Using a set of true first order logic clauses as
base
◦ Has the advantage of knowledge presentation
in logic
◦ Program “What” instead of “How” (like in C)

Biomedical Dept, NIT Raipur 10/11/09 39


 Horn clauses
◦ First order logic: a clause is a set of positive and
negative literals (atoms and atom negation)
◦ Horn clauses: a maximum of one positive literal
per clause
◦ Using Horn clauses decreases expressiveness but
considerably improves efficiency

Biomedical Dept, NIT Raipur 10/11/09 40


 Horn clauses
◦ Number of atoms in the head is 0 or 1
◦ Number of atoms in the body is 0 or more
◦ Empty clause: both head and body have 0
clauses
 Procedural meaning
◦ The procedural call of the program
 Declarative meaning
◦ Logical meaning

Biomedical Dept, NIT Raipur 10/11/09 41


Horn clause Procedural Declarative
meaning meaning
fact A. Definition of A is TRUE
procedure A
rule A :- B1, B2, …, Bn For invoking the IF B1, B2, …, Bn
procedure A, the THEN A
procedures
B1, B2, …, Bn
have to be called in
order

query :- B1, B2, …, Bn Start the calculationNegation of


B1, B2, …, Bn
empty clause □ Termination of the Contradiction
calculation
Biomedical Dept, NIT Raipur 10/11/09 42
 PROLOG uses proof by contradiction
◦ SLD resolution: unification of atoms in the body
of goal clause with the heads of unit clauses or
definite clauses
◦ Unification: manipulating two predicates to
make them appear the same
◦ If there are no candidates for unification,
backtrack to the previous goal and try a
different unification candidate
◦ If the empty clause can be derived, the program
terminates successfully

Biomedical Dept, NIT Raipur 10/11/09 43


 not surprisingly, early semantic nets did not
scale well
◦ most links were general associations
◦ no real basis for structuring semantic relations
 much research has been done in defining
richer sets of links
◦ rely on richer formalism, not richer domain
knowledge

Biomedical Dept, NIT Raipur 10/11/09 44


 Conceptual Dependency theory:
◦ primitives of meaning
1. Actions
2. Objects
3. modifiers of actions
4. modifiers of objects

Biomedical Dept, NIT Raipur 10/11/09 45


 Conceptual Dependency Theory:
1. Actions
1. transfer a relationship (give)
2. transfer physical location of an object (go)
3. apply physical force to an object (push)
4. move body part by owner (kick)
5. grab an object by an actor (grasp)
6. ingest an object by an animal (eat)
7. expel from an animal’s body (tell)
8. transfer mental information (decide)
9. conceptualize or think about an idea (think)
10. produce sound (say)
11. focus sense organ (listen)

Biomedical Dept, NIT Raipur 10/11/09 46


 Conceptual Dependency theory:
primitives of meaning
1. Actions
2. Objects
3. modifiers of actions
4. modifiers of objects
conceptual syntax rules
 built using these primitives
 constitute a grammar of meaningful semantic
relationships.
conceptual dependency relationships
 are defined using the conceptual syntax rules
 can be used to construct an internal representation of an
English sentence.

Biomedical Dept, NIT Raipur 10/11/09 47


◦ Tense and mode are added.
 Example:
 past
 future
 transition
 etc.

Biomedical Dept, NIT Raipur 10/11/09 48


Biomedical Dept, NIT Raipur 10/11/09 49
Biomedical Dept, NIT Raipur 10/11/09 50
 Conceptual Dependency Theory (Schank, 1973)
 attempts to model the semantic structure of natural language
 4 primitive conceptualizations, from which meaning is built

ACT action
PPobjects (picture producers)
AA modifiers of actions (action aiders)
PA modifiers of objects (picture aiders)

primitive actions include: ATRANS (transfer a relationship, e.g.,


give)
PTRANS (transfer physical location, e.g., move)
MTRANS (transfer mental information, e.g., tell)
...

Biomedical Dept, NIT Raipur 10/11/09 51


 a script is a structure that describes a
stereotyped sequence of events in a
particular context
◦ closely resembles a frame, but with additional
information about the expected sequence of
events and the goals/motivations of the actors
involved

◦ the elements of the script are represented using


Conceptual Dependency relationships (as such,
actions are reduced to conceptual primitives)

Biomedical Dept, NIT Raipur 10/11/09 52


 EXAMPLE: restaurant script
describes: items usually found in a restaurant
people and their roles (e.g., chef, waiter, …)
preconditions and post conditions
common scenes in a restaurant: entering,
ordering, eating, leaving
 props and roles are
identified

 pre- and post-


conditions

 CDs describe actions


that occur in each of
the individual scenes

Biomedical Dept, NIT Raipur 10/11/09 53


Biomedical Dept, NIT Raipur 10/11/09 54

S-ar putea să vă placă și