Scribd
Upload
18 views

Chapter 2

Uploaded by

humaira'
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
18 views

Chapter 2

Uploaded by

humaira'
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 22

DAT21003

Artificial Intelligence
Knowledge Representation
What is Knowledge Representation?
 Knowledge representation (KR) is concerned with
intelligent agents thinking, and how thinking contributes
to intelligent behaviour of agents.
 It is responsible for representing information in a way a
computer can understand, and utilize the knowledge to
solve problems.
 It also defines how knowledge is represented in AI.
What is Knowledge Representation?

 The designer of an AI system must:


 Represent the problem in a language the computer can reason.
 Use the computer to compute an output (i.e. answer to be
presented / sequence of actions to be carried out).
 Interpret the output as a solution to the problem.
The Different Kinds of Knowledge
 Declarative knowledge
 To know about something.
 It
represents the facts, objects, concepts that help describe the
world.
 Procedural knowledge
 More complex than declarative knowledge.
 Thisknowledge is used to accomplish any task using certain
procedures, rules, and strategies, making the system that use this
knowledge work efficiently.
The Different Kinds of Knowledge
 Meta-knowledge
 Type of knowledge when combined.
 It’s the knowledge regarding other types of knowledge.
 Heuristic knowledge
 The knowledge provided by experts of certain domains, subjects,
disciplines, and fields.
 Helps in taking the best approach to problems and making decisions.
 Structural knowledge
 Basic form of knowledge to solve real-world problems.
 It describes the relationship that exists between concepts or objects.
Knowledge Representation
 There are multiple approaches about representation.
 Example:
 Pictures and symbols
 Graphs and network
 Numbers
Knowledge Representation
 Pictures and symbols
 Pictorial representation are not easily translated to useful
information for computer because computer can’t interpret
pictures directly without complex reasoning.
 Graphs and network
 Procedural knowledge can be represented using graphs.
 Numbers
 Numbers are an integral part of knowledge representation used
by humans.
 Numbers representation also can be translated easily to
computer.
Cycle of knowledge representation
AI system has the following components for displaying
intelligent behavior:
Relation between knowledge and intelligence
 Knowledge of real-worlds plays a vital role in intelligence
and same for creating artificial intelligence.
 Knowledge plays an important role in demonstrating
intelligent behavior in AI agents.
 An agent is only able to accurately act on some input
when he has some knowledge or experience about that
input.
Knowledge Representation Techniques
1. Logical representation
2. Production rules
3. Semantic networks representation
4. Frame representation
Knowledge Representation Techniques
1. Logical representation

 Logic is based on truth.


 It represents a conclusion based on various conditions and set some
important communication rules.
 Can be categorized into two logics:
 Propositional logics
 First order predicate logics
 Each sentence can be translated into logic using syntax and
semantic.
Knowledge Representation Techniques
1. Logical representation

 Syntax
 It decides how we can construct legal sentences in
logic (like the grammar of a language).
 It determines which symbol we can use in
knowledge representation and how to write those
symbols.
 Semantics
 The rules by which we can interpret the sentence
in the logic.
 It assigns a meaning to each sentence.
Knowledge Representation Techniques
1. Logical representation

 Advantages
 Helps to perform logical reasoning.
 It is the basis for the programming languages.
 Disadvantages
 It has some restrictions and are challenging to work with.
 It may not be very natural, and inference may not be very
efficient.
Knowledge Representation Techniques
2. Production rules

 The agent checks for the condition.


 If the condition exists, then production rule fires and
corresponding action is carried out.
 The condition part of the rule determines which rule may
be applied to a problem.
 The action part carries out the associated problem-solving
steps.
 This complete process is called a recognize-act cycle.
Knowledge Representation Techniques
2. Production rules

 Advantages
 Production rules are expressed in natural language.
 Production rules are highly modular and can be easily
removed or modified.
 Disadvantages
 It does not exhibit any learning capabilities and does not
store the result of the problem for future uses.
 During the execution of the program, many rules may be
active. Thus, rule-based production systems are inefficient.
Knowledge Representation Techniques
3. Semantic network representation
 Work as an alternative of predicate logic.
 It represents knowledge, in the form of
graphical networks.
 This network consists of nodes,
representing objects and arcs, which
describe the relationship between those
objects.
 It can categorize the object in different
forms and can also link those objects.
 This representation consist of two types
of relations:
 IS-A relation (inheritance)
 Kind-of-relation
Knowledge Representation Techniques
3. Semantic network representation
 Advantages
 A natural representation of knowledge.
 It conveys meaning in a transparent manner.
 These networks are simple and easy to understand.
 Disadvantages
 Take more computational time at runtime.
 These are inadequate as they do not have any equivalent quantifiers.
 These networks are not intelligent and depend on the creator of the
system.
Knowledge Representation Techniques
4. Frame representation

 A frame is a record like structure that consists of a collection of


attributes and values to describe an entity in the world.
 These are the AI data structure that divides knowledge into
substructures by representing stereotypes situations.
 Basically, it consists of a collection of slots and slot values of any
type and size.
 Slots have names and values which are called facets.
Knowledge Representation Techniques
4. Frame representation

An Example of a Frame representation

Slots Filter
Name Peter
Profession Doctor
Age 25
Weight 78
Knowledge Representation Techniques
4. Frame representation

 Advantages
 It makes programming easier by grouping related data.
 Easy to understand and visualize.
 Very easy to add slots for new attributes and relations.
 Easy to include default data and search for missing values.
 Disadvantages
 In frame system inference, the mechanism cannot be easily processed.
 The inference mechanism cannot be smoothly proceeded by frame
representation.
 It has a very generalized approach.
Representation Requirements
 Representational Accuracy
 It should represent all kinds of required knowledge.
 Inferential Adequacy
 It should be able to manipulate the representational
structures, to produce new knowledge, corresponding to the
existing structure.
Representation Requirements
 Inferential Efficiency
 The ability to direct the inferential knowledge mechanism,
into the most productive directions, by storing appropriate
guides.
 Acquisitional Efficiency
 The ability to acquire new knowledge easily, using automatic
methods.

You might also like