KNOWLEDGE

REPRESENTATION ISSUES
KNOWLEDGE
Knowledge is a general term.
An answer to the question, "how to represent knowledge", requires an
analysis to distinguish between knowledge “how” and knowledge
“that”.
■ knowing "how to do something".
e.g. "how to drive a car" is Procedural knowledge.
■ knowing "that something is true or false".
e.g. "that is the speed limit for a car on a motorway" is Declarative
Knowledge

2
TWO DIFFERENT KIND OF ENTITIES:

 Facts : truths in some relevant world. Things we want to

represent.

 Representation of facts in some chosen formalism. Things we

will be able to manipulate.

3
KNOWLEDGE IS A GENERAL TERM.

 Knowledge is a progression that starts with data which is

of limited utility.
 By organizing or analyzing the data, we understand

what the data means, and this becomes information.

 The interpretation or evaluation of information yield

knowledge.
 An understanding of the principles embodied within the

knowledge is wisdom.

4
KNOWLEDGE PROGRESSION:

5
CONT….
 Data is viewed as collection of disconnected facts.
Example : It is raining.

 Information emerges when relationships among facts are established and

understood; Provides answers to "who“, "what", "where", and "when“.
Example : The temperature dropped 15 degrees and then it started raining.

Knowledge emerges when relationships among patterns are identified and

understood; Provides answers as "how" .
 Example : If the humidity is very high and the temperature drops substantially, then
atmospheres is unlikely to hold the moisture, so it rains.

 Wisdom is the pinnacle of understanding, uncovers the principles of

relationships that describe patterns. Provides answers as "why" .:
Example : Encompasses understanding of all the interactions that happen between raining,
evaporation, air currents, temperature gradients, changes, and raining.
6
KNOWLEDGE MODEL

7
CONT…
 Data and information deal with the past; they are based
on the gathering of facts and adding context.

 Knowledge deals with the present that enable us to

perform.

 Wisdom deals with the future, acquire vision for what

will be, rather than for what is or was.

8
KNOWLEDGE REPRESENTATION

9
Once acquired, knowledge
must be organized for use
 KNOWLEDGE
REPRESENTATION
Importance of Knowledge

Knowledge can be defined as the body of facts and principles

accumulated by human-kind or the act, fact or state of knowing.
In actuality it is more than this, it also includes having a familiarity with
language, concepts, procedures, rules, ideas, abstractions , places,
customs, facts associations along with ability to use these notions
effectively in modeling different aspects of the world.

10
HOW TO STORE KNOWLEDGE
 In biological organisms
 Human brain weighs 3.3 pounds - estimated number of neurons 10 12 -
potential storage – 1014
 Estimated 1000 billion neurons in the human brain, with each connected
to up to 10 thousand others
 brain loses over 50,000 neurons each day, yet most people do not feel
as though a large part of their memory disappeared
 Electrical impulses produced by a neuron travel along the axon
 The axon connects to dendrites through synaptic junctions

11
AN OVERVIEW OF THE
BIOLOGICAL NEURON

12

ICT619 S2-05 Photo: Osaka University

THE SYNTHETIC NEURON
 A simplemodel of the biological neuron, first
proposed in 1943 by McCulloch and Pitts consists of
a summing function with an internal threshold, and
"weighted" inputs as shown below.

13
HOW TO STORE KNOWLEDGE
 In computers, knowledge is also stored as symbolic structures, in the
form of collections of magnetic spots and voltage states.
 Magnetic storage
 Optical storage
 Spintronics ("spin transport electronics“)

14
KNOWLEDGE TYPE
Tacit knowledge Explicit knowledge

 Exists within a human  Exists outside a human

being; being;
 it is embodied  it is embedded.
 Difficult to articulate  Can be articulated formally
formally  Can be shared, copied,
 Difficult to processed and stored.
share/communicate  Easy to steal or copy
 Hard to steal or copy.  Drawn from artifact of some
 Drawn from experience, type as principle, procedure,
action,subjective insight. process, concepts. 15
16
KNOWLEDGE TYPE
Procedural knowledge Declarative knowledge

 focuses on tasks that must be  .refers to representations of

performed to reach a particular objects and events; knowledge
objective or goal. about facts and relationships;
 Knowledge about "how to do  Knowledge about "that
something"; e.g., to determine somethingis true or false". e.g.,
if Peter or Robert is older, first A car has four tyres; Peter is
find their ages. older than Robert;
 examples : procedures,  example : concepts, objects,
rules, strategies, agendas, facts, propositions, assertions,
semantic nets, logic and
models. descriptive models
17
KNOWLEDGE
REPRESENTATION
 Problem solving requires large amount of knowledge and
some mechanism for manipulating that knowledge.
 The Knowledge and the Representation are distinct

entities, play a central but distinguishable roles in

intelligent system.
 − Knowledge is a description of the world;

 − Representation is the way knowledge is encoded;

 In simple words, we :

− need to know about things we want to represent , and

− need some means by which things we can manipulate.
18
CONT…
 Knowledge representation can be considered at two
levels :
(a) knowledge level at which facts are described, and
(b) symbol level at which the representations of the
objects, defined in terms of symbols, can be manipulated
in the programs.

 A good representation enables fast and accurate access to

knowledge and understanding of the content.

19
 KNOWLEDGE
REPRESENTATION
Let’s take the following examples
Amitabh Bachhan is tall.

Chandra Prakash
X Loves Y.
Mohit has learned to use recursion to manipulate linked lists in several
programming languages.

1st one represents a fact, an attribute possessed by a person.

2nd expresses a complex binary relation between two persons.
3rd is most complex, expressing relations between a person and more
abstract programming concepts.
20
 KNOWLEDGE
REPRESENTATION
Knowledge can be of following types
Declarative (statements)

Chandra Prakash
Procedural (facts)
Heuristics (rule of thumb / experience)
We should not confuse Knowledge with data.
Epistemology – study of nature of knowledge.
Metaknowldge – knowledge about knowledge.

21
 FACTS, HYPOTHESIS AND BELIEF
 Facts:
 Statement may be true or False.
 Eg: Apple falls from a tree.

Chandra Prakash
 Hypothesis:
 Hypothesis is a justified belief that is not known to be true.
 Eg: Newton proposes a Hypothesis that “Apple falls from a tree
because of gravity ”

 Belief or Rule:
 Belief is any meaningful and coherent expression that can be
represented.
 Eg: netwon proves this hypothesis on the basis of certain facts.
22
MAPPING BETWEEN FACTS AND
REPRESENTATION

 Knowledge is a collection of “facts” from some domain.

 We need a representation of facts that can be manipulated by a program.
 Normal English is insufficient, too hard currently for a computer program
to draw inferences in natural languages. Thus some symbolic
representation is necessary.
 Therefore, we must be able to map "facts to symbols" and "symbols to
facts" using forward and backward representation mapping.

23
MAPPING BETWEEN FACTS AND
REPRESENTATION

 Consider an English statement : Spot is a dog

24
25
FORWARD AND BACKWARD
REPRESENTATION
 The forward and backward representations are elaborated below :

 The doted line on top indicates the abstract reasoning process that a
program is intended to model.
 The solid lines on bottom indicates the concrete reasoning
process that the program performs.
26
SIX BASIC
CHARACTERISTICS:

 Coverage, which means the KR covers a breath and depth of information. Without
a wide coverage, the KR cannot determine anything or resolve ambiguities.
 Understandable by humans. KR is viewed as a natural language, so the logic
should flow freely. It should support modularity and hierarchies of classes (Polar
bears are bears, which are animals). It should also have simple primitives that
combine in complex forms.
 Consistency. If John closed the door, it can also be interpreted as the door was
closed by John. By being consistent, the KR can eliminate redundant or
conflicting knowledge.
 Efficient
 Easiness for modifying and updating.
 Supports the intelligent activity which uses the knowledge base

27
KNOWLEDGE
REPRESENTATION
A good system for the representation of knowledge in a particular domain
should possess the following properties :

•Representational Adequacy
•Ability to represent all of the kinds of knowledge that are needed in that
domain
•Should collect all related facts
•Inferential Adequacy
•Ability to manipulate the representational structures in such a way as to
derive new structures corresponding to new knowledge inferred from old.
•All rules
28
CONT…
• Inferential Efficiency
• The ability to incorporate into the knowledge structure
additional information that can be used to focus the
attention of the inference mechanisms in the most
promising direction.
• Best available rule
• Acquisitional Efficiency
• Acquiring new information easily

• * Till date No single system can optimizes all of the above

29
properties
KNOWLEDGE REPRESENTATION
SCHEMES :
There are four types of Knowledge representation :

 Relational Knowledge

 Inheritable Knowledge

 Inferential Knowledge

 Declarative / Procedural Knowledge

30
KNOWLEDGE
REPRESENTATION
1. Simple Relational Knowledge
Relational knowledge is made up of objects consisting of attributes and
their corresponding associated values.
− The results of this knowledge type is a mapping of elements among different

Chandra Prakash
domains.
Player Height Weight Runs Bats-
Throws
X 6-0 150 3000 Right-
right
Y 5-4 140 15000 Left-
right
Z 6-2 160 1000 Right-
left

ThisPlayer_info(‘X’, ‘6-0’,
even does not tell ‘3000’,
us that who is‘Right-
the longest player? 31
right’)
A procedure should be defined to figure out the result.
KNOWLEDGE
REPRESENTATION
2. Inheritable Knowledge
It is possible to enhance the basic representation with inference
mechanism that operate on the structure of the representation.

Chandra Prakash
The most useful form of inference is Property Inheritance, in which
elements of specific classes inherit attributes and values from ore general
classes in which they are included.
In order to do this, objects must be organized into classes and classes
must be arranged into generalization hierarchy.
Cricketer
Isa : adult-Male
Bats : handed
Height ; 6-1 etc. 32
Property Inheritance is what happens in Inheritable Knowledge
INHERITABLE KNOWLEDGE
Semantic network representation
Frames based representation

Chandra Prakash
33
34
 the directed arrows represent attributes (isa,
instance, and team) originating at the object
being described and terminating at the object or
its value.
 the box nodes represents objects and values of

the attributes.

35
SEMANTIC NETWORK EXAMPLE

Is a Human
Boy Being

36
a
a Is
Is
Goes to Needs
Woman
School Joe
a
Is Food
Has
a child
Kay
HOW FRAMES ARE ORGANISED
 The idea of frame hierarchies is very similar to the idea of class hierarchies
found in object-orientated programming
 A frame system is a hierarchy of frames
 Each frame has:
 a name.

 slots: these are the properties of the entity that has the name, and they have
values. A particular value may be:

 a default value
 an inherited value from a higher frame
 a procedure, called a daemon, to find a value
 a specific value, which might represent an exception.

37
VIEWING A NODE AS A FRAME:
Baseball-player
isa : Adult-Male
Bates : EQUAL handed
Height : 6.1
Batting-average : 0.252

38
INFERENTIAL KNOWLEDGE :
 Generates new information from the given information.
This new information does not require further data
gathering form source, but does require analysis of the
given information to generate new knowledge.
 Given a set of relations and values, one may infer other values or
relations.
 a predicate logic (mathematical deduction) is used to infer from
a set of attributes.
 Inference through predicate logic uses a set of logical operations
to relate individual data. The symbols used for the logic
operations are :
" → " (implication), " ¬ " (not), " V " (or), " Λ " (and), " ∀ "
(for all), " ∃ " (there exists). 39
EXAMPLES OF PREDICATE
LOGIC STATEMENTS :
1. Wonder is a name of a dog : dog (wonder)
2. Tommy is a name of a dog: dog(tommy)
3. All dogs belong to the class of animals : ∀ x : dog
(x) → animal(x)
4. All animals either live on land or in water : ∀ x :
animal(x) → live (x, land) V live (x, water)

 We can infer from these three statements that :

" Wonder lives either on land or on water."
 As more information is made available about these
40
objects and their relations, more knowledge can be
inferred.
DECLARATIVE/PROCEDURAL
KNOWLEDGE :
 Declarative knowledge :
Here, the knowledge is based on declarative facts about
axioms and domains.
 axioms are assumed to be true unless a counter
example is found to invalidate them.
 domains represent the physical world and the
perceived functionality.
 axiom and domains thus simply exists and serve as
declarative statements that can stand alone.

41
 CONT…
 Procedural Knowledge
The previous forms deal with Static, Declarative facts.
 This knowledge specifies what to do and when.
This knowledge can be represented in programs in many ways. The most
common way is simply as code for doing something.
The machine uses the knowledge when it executes the code to perform a
task.
Eg : If – then else etc.

The procedural knowledge are represented as small programs that know how to do
specific things, how to proceed.
42
ISSUES IN KNOWLEDGE
REPRESENTATION:
• Important Attributes : Any attribute of objects so basic that they occur in
almost every problem domain ?

• Relationship among attributes: Any important relationship that exists

among object attributes ?

• Choosing Granularity : At what level of detail should the knowledge be

represented ?

• Set of objects : How sets of objects be represented ?

• Finding Right structure : Given a large amount of knowledge stored, how

can relevant parts be accessed ?
43
KNOWLEDGE
REPRESENTATION
Important Attributes :
instance and isa
Relationship among attributes :
Inverses
This is about consistency check, while a value is added to one attribute .

Existence in an isa hierarchy:

height is a specialization of physical-attributes
Techniques for reasoning about values:
height must be no .
Age of child cant be greater than parents.
Single-values attributes 44

A member can be in team at a time.

CHOOSING GRANULARITY :
 At what level should the knowledge be represented and what are the
primitives.
 Should there be a small number of High-level facts or should there
be a large number of low-level primitives.
 High-level facts may not be adequate for inference while Low-level
primitives may require a lot of storage.

45
EXAMPLE OF GRANULARITY
:
 Suppose we are interested in following facts:
John spotted Tom.
 This could be represented as
Spotted (agent(John), object (Tom))
 Such a representation would make it easy to answer questions such are :
Who spotted Tom?
 Suppose we want to know :
Did John see tom?
− Given only one fact, we cannot discover that answer.
− We can add other facts, such as
Spotted (x , y) → saw (x , y)
 We can now infer the answer to the question
46
SETS OF OBJECT
 There are certain properties of objects that are true as member of a set but
not as individual;
 The reason to represent sets of objects is : If a property is true for all or
most elements of a set, then it is more efficient to associate it once with the
set rather than to associate it explicitly with every elements of the set . This
is done,
 in logical representation through the use of universal quantifier, and
 in hierarchical structure where node represent sets and inheritance propagate set
down to individual.
 However in doing so, for example: assert large (elephant), remember to
make clear distinction between,
 whether we are asserting some property of the set itself, means, the set of elephants

is large

47
FINDING RIGHT STRUCTURE :
 This is about access to right structure for describing a
particular situation.

 This requires, selecting an initial structure and then

revising the choice.

48