Transcript What is a symbol? - UC Berkeley School of Information

Introduction to Knowledge
Representation
Marti Hearst
SIMS 202: Information Organization
and Retrieval
Lecture 6, Sept 10, 1998
Today




What is a symbol?
Semantics: the meanings of symbols
Creating Ontologies
Objects, Properties, and Relations
References:
Chapter 1 of Introduction to Knowledge Systems by Mark
Stefik.
Chapter 8 of Artificial Intelligence, A Modern Approach by
Stuart Russell and Peter Norvig
What is a symbol?

From Merriam-Webster’s Collegiate:
Something that stands for or suggests
something else.
 An arbitrary conventional sign used in
writing or printing to represent:

operations
 quantities
 elements
 relations
 qualities


What is meant by sign and represent?
Recognizing Symbols

What is/are this/these symbol(s)?
Two overlapping squares?
 Eight horizontal and eight vertical lines?

Designation and Representation


What does it mean to represent something?
Identify the following:
symbols
 context
 observer



Representation is the association of symbols
with conceptual objects or ideas in a given
context.
The observer sets up a correspondence
between the symbols and the meanings.
Representation with Symbols
Kailin threw the ball to Juno.
did-action
Juno
thrown-to
Kailin
throw event
objectthrown
thrower
a ball
Symbols and Language


Abstract concepts are difficult to express in a
computer.
Combinations of abstract concepts are even
more difficult to express:
time
 shades of meaning
 social and psychological concepts
 causal relationships

Symbols and Language
The Dog.
Symbols and Language
The Dog.
The dog cavorts.
The dog cavorted.
The picture doesn’t really show the manner or tense.
Symbols and Language
The man.
The man walks.
Symbols and Language
The man walks the cavorting dog.
So far, we can sort of show the meaning in pictures.
Symbols and Language
As the man walks the cavorting dog, thoughts
arrive unbidden of the previous spring, so unlike
this one, in which walking was marching and
dogs were baleful sentinals outside unjust halls.
What is the relation between the symbols and the meaning?
Symbols and Language


Language only hints at meaning.
Most meaning of text lies within our minds
and common understanding.

“How much is that doggy in the window?”
how much: social system of barter and trade
(not the size of the dog)
 “doggy” implies childlike, plaintive, probably
cannot do the purchasing on their own
 “in the window” implies behind a store window,
not really inside a window, requires notion of
window shopping

Setting up Correspondences
between Symbols and Meaning

Consider made-up languages

Codes used by espionage agents
“Pope” means a particular piece of microfilm
 “Denver” indicates a particular mailbox


People remember the “gist” instead of the
actual words used.

This implies the actual words used are not
very salient; what matters is the meaning.
Recognizing Symbols


The marks that constitute a symbol depend
on the conventions for recognizing the
symbol.
A recognizer typically has an associated
alphabet or set of symbols:
Token: an individual instance of a symbol
 Type: a class of symbols


Examples?
The Role of Context

The concept associated with the symbol 21
means different things in different contexts.


Examples?
The question “Is there any salt?”
Asked of a waiter at a restaurant.
 Asked of an environmental scientist at
work.

Semantics: The Meaning of
Symbols


Semantics versus Syntax
Meaning versus Representation

What a person’s name is versus who they
are.


A rose by any other name...
What the computer program “looks like”
versus what it actually does.
Semantics

Semantics: assigning meanings to
symbols and expressions.

Usually involves defining:
objects
 properties of objects
 relations between objects


More detailed versions include (among
others)
events
 time
 places
 measurements (quantities)

Ontology

From Merriam-Webster’s Collegiate:
A branch of metaphysics concerned with
the nature and relations of being.
 A particular theory about the nature of
being or the kinds of existence.


More prosaically:
A carving up of the world’s meanings.
 Determine what things exist, but not how
they inter-relate.


Related terms:

taxonomy, dictionary, category structure
Knowledge Engineering Steps




Decide what to talk about
Decide on a vocabulary of predicates,
functions, and constants
Encode general knowledge about the domain
Artificial Intelligence vs Cataloging:
AI goal: allow computer programs to
reason about the objects and relations
 Cataloging: organize the objects and
relations for use by humans
 AI is more ambitious and more difficult
 We aren’t covering the reasoning part
here.

Try some examples

Let’s define
Types of Objects
 Types of Properties of Objects
 Types of Relations between Objects

Attributes vs. Objects


How do we make this distinction?
Say we are clothing manufacturers.
Fur is a class of objects
 Animal is an attribute of this class


Say we are naturalists.
Animal is a class of objects
 Fur is an attribute of this class

Garment Maker Ontology

Define the objects; Indicate what types of attributes
are used to define the objects (attributes=properties)

Object Class:


Attribute Types:





ISA:
Material:
Color:
Garment_Type:
Object Class:


Garment
Fur
Attribute Types:




ISA:
Animal:
Color:
Texture:
Garment Maker Ontology

Attributes have lists of legal values

Object Class: Garment





ISA: Object
Material: fur, cotton, wool
Color: red, black, brown, white, blue
Garment_Type: coat, stole, hat
Object Class: Fur




ISA: Material
Animal: fox, rabbit, sable
Color: red, black, white
Texture: silky, thick, coarse
Garment Maker Ontology

Show the assignments of values to attributes for one
particular instance of an object

Object Class: Garment




ISA: Object
Material: fur, cotton, wool
Color: red, black, brown, white, blue
Garment_Type: coat, stole, hat
object
ISA
garment
Material

Object Class: Fur




fur
ISA: Material
Animal: fox, rabbit, sable
Animal Color
Color: red, black, white
red
Texture: silky, thick, coarse sable
G_type
coat
Texture
silky
Garment Maker Ontology


Usually only one value is allowed for an ISA attribute
In this example,


The value of the color attribute for Garment is
determined by the color attribute for the garment’s
Material attribute
object
ISA
This is called inheritance
garment
Material
G_type
fur
Animal
sable
Color
red
coat
Texture
silky
Garment Makers vs. Naturalists

A difference between a class definition and an
attribute value

Class Fur
ISA: material
 Animal: fox, rabbit, sable
 Color: red, black, white
 Texture: silky, thick, coarse
 Garment_type: coat, stole, hat


Class Animal
ISA: mammal
 Outer_Covering: fur, skin, scales
 Number_of_limbs: 4, 6, 8
 Circulatory_System: cold_blooded,

Nesting Attributes and Classes

Class Garment
 Material:



Class Fur
 Animal: fox, rabbit, sable
 Color: red, black, white
 Texture: silky, thick, coarse
Class Cotton
 Color: red, blue, white, brown, black
 Thread_Count: 100, 200
Garment_type: stole, coat, hat, t-shirt
Attributes often must be nested
 Alternative: two subclasses of Garment

Next Week




Semantic Nets
Facets vs. Hierarchies
Lexical Semantics
Word Associations