INYM 325/329 MEC

brewerobstructionΤεχνίτη Νοημοσύνη και Ρομποτική

7 Νοε 2013 (πριν από 3 χρόνια και 9 μήνες)

80 εμφανίσεις

ARTIFICIAL INTELLIGE
NCE

STUDY GUIDE FOR


INYM

325
/
329

MEC

*
INYM325
/
329
MEC
*

FACULTY OF

COMMERCE AND ADMINISTRATION

MAFIKENG CAMPUS





ii

























Study guide compiled by:

Mthulisi Velempini




Instructional Design by Mrs Annelize Cronje, Academic Staff Advisor, ADC

Page layout by
Roxanne Bremner,

Academic
Development

Centre

Printing arrangements and distribution by Department
Logistics (Distribution Centre).

Printed by.

Nashua Digidoc Centre (018) 299 2827

Copyright


20
13

edition. Date of revision 20
15
.

North
-
West University,
Mafikeng

Campus.

No part of this book may be reproduced in any form or by any means without written
permission from the publisher.



iii

TABLE OF

CONTENTS

Module Information

................................
................................
................................
................

v

Study Guide Title
:
Artificial Intelligence

................................
................................
..................

v

Module Qualification:

Bachelor
of
Commerce
in
Information Systems

................................
...

v

Module Structure:

................................
................................
................................
..................

v

Contact Person:

................................
................................
................................
.....................

v

A) exit level outcomes

................................
................................
................................
............

v

B) critical cross field outcomes:

................................
................................
..............................

v

Study references:

................................
................................
................................
..................

vi

Assessment structure:

................................
................................
................................
..........

vi

Structure of the module

................................
................................
................................
.........

vii

Course outline
................................
................................
................................
.......................

vii

Study icons

................................
................................
................................
..........................

viii

Warning against plagiarism

................................
................................
................................
..

viii

Study u
nit 1:

Introduction
to artificial intelligence

................................
...................

1

1.1

Unit
outcomes

................................
................................
........................

1

1.2

Unit
overview

................................
................................
.........................

1

1.3

Unit
readings

................................
................................
..........................

2

1.4

Important
unit concepts

................................
................................
..........

2

1.5

Assessment
questions

................................
................................
...........

2

Study unit 2:

Natural
language understanding

................................
........................

3

2.1

Unit
outcomes

................................
................................
........................

3

2.2

Unit
overview

................................
................................
.........................

3

2.3

Unit
readings

................................
................................
..........................

4

2.4

Important
unit concepts

................................
................................
..........

4

2.5

A
ssessment questions

................................
................................
...........

5

Study unit 3:

Machine
learning

................................
................................
..................

7

3.1

Unit
outcomes

................................
................................
........................

7

3.2

Unit
overview

................................
................................
.........................

7

3.3

Unit
readings

................................
................................
..........................

8

3.4

Important
unit concepts

................................
................................
..........

8

3.5

Assessment
questions

................................
................................
...........

9

S
tudy unit 4:

Robotics

................................
................................
.............................

11

4.1

Unit
outcomes

................................
................................
......................

11

4.2

Unit
overview

................................
................................
.......................

11

4.3

U
nit readings

................................
................................
........................

12

4.4

Important
unit concepts

................................
................................
........

12


iv

4.5

Assessment
questions

................................
................................
.........

13

Study unit

5:

Reasoning
................................
................................
...........................

15

5.1

Unit
outcomes

................................
................................
......................

15

5.2

Unit
overview

................................
................................
.......................

15

5.3

Unit
readings

................................
................................
........................

16

5.4

Important
unit concepts

................................
................................
........

16

5.5

Assessment
questions

................................
................................
.........

17

Study unit

6:

Decision
making (search, planning, decision theory)

.....................

19

6.1

Unit
outcomes

................................
................................
......................

19

6.2

Unit
overview

................................
................................
.......................

19

6.3

Unit
readings

................................
................................
........................

20

6.4

Important
unit concepts

................................
................................
........

20

6.5

Assessment
questions

................................
................................
.........

20

Study unit

7:

Expect
systems

................................
................................
..................

21

7.1

Unit
outcomes

................................
................................
......................

21

7.2

Unit
overview

................................
................................
.......................

21

7.3

Unit
readings

................................
................................
........................

21

7.4

Important
unit concepts

................................
................................
........

22

7.5

Assessment
questions

................................
................................
.........

2
3

Study unit

8
:

Atrificial
neural networks

................................
................................
..

25

8.1

Unit
outcomes

................................
................................
......................

25

8.2

Unit
overview

................................
................................
.......................

25

8.3

Unit
readings

................................
................................
........................

26

8.4

Important
unit concepts

................................
................................
........

26

8.5

Assessment
questions

................................
................................
.........

27

Study unit

9:

Intelligent
systems

................................
................................
.............

29

9.1

Unit
outcomes

................................
................................
......................

29

9.2

Unit
overview

................................
................................
.......................

29

9.3

Unit
readings

................................
................................
........................

30

9.4

Important
unit concepts

................................
................................
........

30

9.5

Assessment
questions

................................
................................
.........

31

Study unit

10
:

Genetic
algorithms

................................
................................
.............

33

10.1

Unit
outcomes

................................
................................
......................

33

10.2

Unit
overview

................................
................................
.......................

33

10.3

Unit
readings

................................
................................
........................

34

10.4

Important
unit concepts

................................
................................
........

34

10.5

Assessment
questions

................................
................................
.........

34

10.6

Unit moderation

................................
................................
....................

35



v

MODULE INFORMATION

STUDY GUIDE TITLE
:
ARTIFICIAL INTELLIGE
NCE

MODULE
QUALIFICATION:

BACHELOR OF COMMERCE

IN
INFORMATION SYSTEMS

MODULE
STRUCTURE:

Credits

NQF Level

Code

Type

Duration

Semester

16

7

INYM

325
/329

Core

16 Weeks

Second

CONTACT PERSON:

Position

Name

Phone and email

Office

Lecturer

Mthulisi
Velempini

018

389 2056

Mthulisi.velempini@nwu.ac.za

A132
General
Academic
Building

A) EXIT LEVEL
OUTCOMES

The learner will be able to:



Define artificial intelligence (AI) giving examples, current state of research, successes
and failures of AI



Describe all the branches of AI and discuss at

least

one example of each branch



Discuss the importance and
the need
for

artificial intelligence



Debate giving facts and supporting idea whether AI can replace human beings



Justify or refute the use of the intelligence

in machines

and whether machines can
exhibit intelligent behaviour



Design an architecture of
neural networks



Implement AI algorithms to solve given problems



Technical competence

B) CRITICAL CROSS FI
ELD OUTCOMES:

Upon completin
g

this module, a learner would have attained the following life skills:



Analytical and problem solving skills.



Managerial a
nd good design techniques



Team building leadership skills



An inquisitive and research inclined mind



Good, effective communication and presentation skills



Ability to solve cross cutting challenges


vi



Time and life management skills



Be ethical and be able to i
dentify with his or her community

Course Description

The module introduces learners to the field of Artificial Intelligence (AI). AI is a broad
computer science field which seek to develop computer hardware and software tools which
exhibit human reasoning
and intelligence. It consists of
a
number of sub areas focused on
specific areas of study.
The most common sub fields are:

Machine Learning, Neural network,
Robotics, Natural language understanding, Knowledge representation,
Intelligent Systems,
Reasoning
and decision making. The learners will be introduced to all the sub fields of AI.
The background information, applications, challenges, failures, research and future directions
of these sub fields will be discussed. However, more emphasis will be put on so
ftware
components of AI in which learners will be required to work with a number of AI related
algorithms and convert them into software systems.

STUDY REFERENCES:

The following study material will be used in the development of this learning module.

Presc
ribed textbook



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Prentice Hall. 2003.

Referral textbooks



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesle
y. 2005.



Hofstadter, DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide to Intelligent Systems, 2/E


Michael Negnevitsky
,

School of Electrical Engineering and Computer
Science, University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660 Publisher:

Addison
-
Wesley

ASSESSMENT STRUCTURE
:

The assessment consists of formative and summative assessments. The following
summative assessments will be assigned to learners: one semester project, three tests, two
assignments and two quizzes.
The
setting techniques
of f
ormative assessments will vary
according to the understanding of learners. Their frequency will also depend on the progress
of learners. Most of the assignments will involve coding in high level programming
languages. The final examination will be written
at the end of the semester.
The following is
the contribution of each type of assessment:



Tests
:






(20%)



Assignments
:




(5%)



Quizzes:





(5%)



Semester Project:



(20%)



Final Examination:



(50%)


T
otal:





100%



vii

STRUCTURE OF THE MOD
ULE

Week

Topic

Unit

Assessment Task

1

Introduction to Artificial Intelligence

Unit 1

Quiz One

2

Natural language understanding

Unit 2

Assignment One

3

Machine Learning

Unit 3

Quiz Two

4

Robotics

Unit 4

Test One

5

Reasoning
(logical, probabilistic)

Unit 5

Quiz Three

6

Decision making
(search, planning, decision theory)

Unit 6

Assignment Two

7

Intelligent Systems

Unit 7

Quiz Four

8

Expect Systems

Unit 8

Test Two

9

Neural network

Unit 9

Quiz Five

10

Genetic algorithms

Unit 10

Assignment Three

11

Perception:
vision, speech understanding

Unit 11

Test Three

12

Review

Unit 12


13

Review

Unit 13

Project Submission

Final Examination

COURSE OUTLINE



Introduction to Artificial Intelligence



Natural language understanding



Machine Learning



Robotics



Reasoning
(logical, probabilistic)



Decision making
(search, planning, decision theory)



Intelligent Systems




Expect Systems



Neural network



Genetic algorithms



Perception: vision, speech understanding


viii

STUDY ICONS


Introductory
statements


Individual exercise


Study the indicated material(s)
in the textbook / article, etc




Outcomes



General overview



List of concepts with or without
explanation


WARNING AGAINST PLAG
IARISM


ASSIGNMENTS ARE INDIVIDUAL TASKS AND NOT
GROUP ACTIVITIES
.
(UNLESS
EXPLICITLY INDICATED AS GROUP ACTIVITIES)

Copying
of text from other learners or from other sources (for instance the study guide,
prescribed material or directly from the internet) is
not allowed



only brief quotations are
allow
ed and then only if indicated as such.

You should
reformulate

existing text and use your
own words
to explain what you have
read. It is not acceptable to retype existing text and just acknowledge the source in a
footnote


you should be able to relate the

idea or concept, without repeating the original
author to the letter.

The aim of the assignments is not the reproduction of existing material, but to ascertain
whether you have the ability to integrate existing texts, add your own interpretation and/or
c
ritique of the texts and offer a creative solution to existing problems.

Be warned: students who submit copied text will obtain a mark of zero for the
assignment and disciplinary steps may be taken by the Faculty and/or University. It is
also unacceptable
to do somebody else’s work, to lend your work to them or to make
your work available to them to copy


be careful and do not make your work available
to anyone!


Study unit 1


1


1

INTRODUCTION TO
ARTIFICIAL INTELLIGE
NCE





1.1

UNIT OUTCOMES

Learners
will be able to:



Describe the sub areas of AI



Trace the history of AI



Describe the foundation and the building blocks of AI



Define key AI terms such as reasoning, knowledge representation, perception, Artificial
and intelligence


1.2

UNIT OVERVIEW

This unit introduces learners to the background and
to
the foundations of AI. It traces the
history of AI and discusses the perceived future directions of AI research. AI
is
categorized
into a number of sub areas and each area is evaluated

in detail
. The u
nit is also define
s

key
terms of AI and discus
se
s the
state of the art of AI

and
its limitations
. The current
applications of AI including current promising AI projects will be discussed.
All the units will
preview the current

AI projects.





Study unit 1


2


1.3

UNIT
READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Prentice Hall. 2003.



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesley. 2005.



Hofstadter, DR,
Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide to Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electrical Engineering and Computer Science,
University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

97
80321204660
Publisher:

Addison
-
Wesley


1.4

IMPORTANT UNIT CONCE
PTS



Understanding AI



What is reasoning



What is intelligence



The Turing test



Foundations of AI



The history of AI



Success and the applications of AI



Sub areas of AI


1.5

ASSESSMENT QUESTIONS



Formative

Assessment



Discuss giving examples and illustrations, the benefits of Artificial Intelligence.



Submit the assignment through e
-
Fundi.



For submission details and due date check e
-
Fundi (Announcement and Assignment
tools)

Study unit 2


3


2

NATURAL LANGUAGE
UNDERSTANDING





2.1

UNIT OUTCOMES

Learners will be able to:



Identify and discuss the areas where natural language processing has been a success
story



Describe the shortcoming of computers in processing natural

language



Discuss the challenges faced by computers in an endeavour to process natural
languages



List and discuss the broad sub divisions of natural language processing



Discuss the benefits of natural language processing



2.2

UNIT OVERVIEW

This unit focuses
on attempts to make computers talk. The unit looks at how computers
process natural language. The interest in natural language processing is highlighted and
areas where natural language processing is required are identified and discussed. The unit
also di
scusses the challenges of natural language processing and the shortcomings of
computers in th
is

sub area of Artificial Intelligen
ce.
The following aspects of natural language
processing are discussed: Syntax, Semantics, Information Extraction, Information
Retrieval,
Machine translation, and Question answering. The areas where natura
l language processing
has been implemented

success
fully

are also discussed.


Study
unit 2


4


2.3

UNIT READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Pr
entice Hall. 2003.



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesley. 2005.



Hofstadter, DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide t
o Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electrical Engineering and Computer Science,
University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660
Publisher:

Addison
-
Wesley



2.4

IMPORTANT UNIT CONCE
PTS



Syntax



Semantics



Information

Extraction



Information Retrieval



Machine translation



Question answering



Natural language recognition



Natural language generation



Speech recognition



Speech generation



Ambiguity



Corpora





Study unit 2


5


2.5

ASSESSMENT QUESTIONS



Summative Assessment



Describe the successfu
l use of NLP in a limited domain




[2]



Identify a Noun Phase and a Verb Phase in the following statements:



Reserve Bank raises interest rates







[1]



I prefer a morning flight









[1]



Describe the challenge likely to be encountered by a natural lan
guage processor in the
following statements:









[1]



The worm was found dead in the plant



I saw a Sow in Saw



What is a Corpora? Discuss all the types of corpora and demonstrate how the tool is
used in natural language pr
ocessing







[5]


Study
unit 2


6



Study unit 3


7


3

MACHINE LEARNING







3.1

UNIT OUTCOMES

Learners will be able to:



Describe general learning and give appropriate examples



Contrast general learning to machine learning



Demonstrate how machines learn from experience



Identify challenges of machine learning and its shortcomings



Identify areas where machine learning is implemented



Compare and contrast the two extreme pattern recognition approaches


discriminative
and generative methods



Demonstrate with an aid of appropr
iate examples how the two approaches of pattern
recognition can be implemented in real world



Identify and discuss current and future directions of machine learning



Characterize and describe the machine learning algorithm development pipeline


3.2

UNIT
OVERVIEW

The unit introduces learners to the concept of machine learning. Furthermore, it describes
how machine through learning
from past experiences
and
pattern
recognition make
s

decisions and inferences. It first introduces learners to learning and ther
eafter contrasts
general learning to machine learning. The importance on machine learning and its benefits
are highlighted. A number of examples are given to help learners to understand the abstract
components of machine learning.

Study unit 3


8


3.3

UNIT READINGS



RUSSELL,
S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Prentice Hall. 2003.



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesley. 2005.



Hofstadter, DR, Gödel, Esher, Bach
: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide to Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electrical Engineering and Computer Science,
University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660
Publis
her:

Addison
-
Wesley



3.4

IMPORTANT UNIT CONCE
PTS



What is learning in general



Defining machine learning



Learning through experience paradigm



Components of machine learning



The significance of machine learning



Application of machine learning



Pattern recogniti
on and how it relates to machine learning



Generative methods



Discriminative methods



The state of art and future research directions of machine learning



The limitations of machine learning






Study unit 3


9


3.5

ASSESSMENT QUESTIONS



Using the given Salmon and the Sea Bass
pictures, apply the generative and the
discriminative methods to identify the differences of

the

two types of fish.



Generative method is a special form discriminative method. Discuss.


Study unit 3


10

Study unit 4


11


4

ROBOTICS







4.1

UNIT

OUTCOMES

Learners will be able to:



Identify and describe basic components of a robot



Draw and label a typical diagram of a robot



Describe the interconnections of a robot system



Describe how different components of a robot interact



Categorize robot
components into hard and software systems


4.2

UNIT OVERVIEW

The unit introduces learners to the exciting field of robotics. It gives a brief background, the
state of art and the future directions of robotics. A complete architecture of robots is
presented

an
d each component is discussed

in details with the aid of diagrams where
possible. Robots are classified according to their design objective
s

and abilities. The need for
robots and the areas where they can be implemented is discussed. The unit answers the
q
uestion, “Why robotics?”





Study unit 4


12


4.3

UNIT READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Prentice Hall. 2003.



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edition.
Addison Wesley. 2005.



Hofstadter, DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide to Intelligent Systems, 2/E


Michael Negnevitsky
,

School of Electrical Engineering and Computer
Science, University of Tasman
ia

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660 Publisher:

Addison
-
Wesley



4.4

IMPORTANT UNIT CONCE
PTS



Robotics definition



History of robots



Characteristics of robots



Components of robots



Brain



Types of Brains



Sensors



Motors



effectors



Chassis


the
frame of the robot


(Body)



Controller



Algorithms and the programming of robots



Programme structure



Types of robots



Mobile



Industrial



Educational

Study unit 4


13



Medical



Entertainment



Classes of robots



State of the art



Laws of robots



Limitations of robots



Why do we need

robots?



Current research



Future research



4.5

ASSESSMENT QUESTIONS



Test One



Are robots smarter than human beings


use e
F
undi discussion forum to debate this
statement



Do a brief research on the improvements on OSIMO robot


link to the latest version of
AS
IMO is available on e
-
fundi


Study unit 4


14

Study unit 5


15


5

REASONING







5.1

UNIT OUTCOMES

Learners will be able to:



Define logic and logical reasoning



Represent a statement in logic form



Construct a proofs to validate a premise



Reason
about content



Define probabilistic reasoning



Define uncertainty



Represent and structure a statement with conditional and dependent variables in a
form of a Bayesian Network



Compare and contrast deductive reasoning to probabilistic reasoning


5.2

UNIT OVERVIEW

Computer or machine reasoning is categorized into two broad divisions: logical and
probabilistic reasoning. Logical reasoning deals with certainty whereas probabilistic
reasoning deals with uncertainty. However, the two reasoning paradigms make use of
dis
crete mathematics in evaluating statements and constructs and to validate premises.
Furthermore, probabilistic reasoning employs probability and Bayesian networks. Teaching
aids and graphical examples are used to explain the two reasoning paradigms.



Study unit 5


16


5.3

U
NIT READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Prentice Hall. 2003.



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesley. 2005.



Hofstadter,
DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide to Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electrical Engineering and Computer Science,
University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:


9780321204660
Publisher:

Addison
-
Wesley



5.4

IMPORTANT UNIT CONCE
PTS



Logical reasoning



Facts and rules



Deductive reasoning



Limits of logical reasoning



Probabilistic reasoning



Dealing with uncertainty



Degree of Belief



Degree of Truth



Handling Uncertain
Knowledge



Basic Probability Theories



Baye`s rule



Probability



Combining evidence



Predictive and diagnostic reasoning



Incorporation of new evidence



conditional probabilities



Evidence

Study unit 5


17



Prior / unconditional probability



Posterior / conditional probability



Uncer
tainty and Rational Decisions



Utility theory



Utility



Decision Theory



Principle of Maximum Expected Utility



Independence



Conditional Independence



Causal Independence



Dependencies



Context
-
specific Dependencies



Asymmetric dependencies



Causality



Inference



Predictive Inference



Bayesian
Networks



Independence assumptions



Product Rule



5.5

ASSESSMENT QUESTIONS



Compare and contrast logical to probabilistic reasoning.



Test two

Study unit 5


18

Study unit 6


19


6

DECISION MAKING (SE
ARCH,
PLANNING, DECISION
THEORY)





6.1

UNIT OUTCOMES

Learners will be able to:



Define basic elements of problems and solutions



Define the
state

space

of the problem



Describe the properties of search strategies



Measure the performance of problem
-
solving



Measure the effectiveness of the search strategy



Generate action sequences



Define a search tree



Discuss using examples at least six search strategies



Describe a general search algorithm



Describe the properties of genetic algorithms


6.2

UNIT OVERVIEW

Introduces searching as a tool for decision making. Artificial Intelligence is all about
searching; hence searching is a fundamental study to AI. All search strategies are
categorized into informed and uniformed taxonomies. The performance of strategies is

compared in terms of space complexity, time
complexity
,
optimality, and completeness.


Study unit 6


20


6.3

UNIT READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Prentice Hall. 2003.



LUGER, GF, Artificial Intelligence Structures
and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesley. 2005.



Hofstadter, DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide to Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electri
cal Engineering and Computer Science,
University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660
Publisher:

Addison
-
Wesley


6.4

IMPORTANT UNIT CONCE
PTS



Basic elements of a problem



Search Strategy



Measuring the performance of the search strategy



M
easuring the effectiveness of the search strategy



Uninformed search strategies



Informed search strategies



Genetic Algorithms


6.5

ASSESSMENT QUESTIONS



Assignment



Compare and contrast any four search strategies





[20]



Test

two

Study unit 7


21


7

EXPECT SYSTEMS








7.1

UNIT OUTCOMES

Learners will be able to:



Describe expert systems (ES) and distinguish them from decision support systems



Identify all the components of ES



Describe with the aid of diagrams the components of ES



Discuss using examples the applications of ES



Discuss using examples the limitations of ES


7.2

UNIT OVERVIEW

This unit gives an overview of Expert systems. Expert systems are intelligent systems
designed to aid management decision making. However, expert sy
stems make decisions on
behalf of management. The components of expert systems will be discussed with the aid of
diagrams. Applications of expert systems will also be presented.


7.3

UNIT READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Appro
ach. Second Edition.
Prentice Hall. 2003.

Study unit 7


22



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesley. 2005.



Hofstadter, DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial
Intelligence: A Guide to Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electrical Engineering and Computer Science,
University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660
Publisher:

Addison
-
Wesley



7.4

IMPORTANT UNIT CONCE
PTS



Expert Systems



Problems with Expert Systems



Some famous expert systems



Structure of expert systems



Features of expert systems



Knowledge representation



Overview of ES



ES Shell



Limitations of ES



Capabilities of ES



Benefits of ES



ES building process



Uses of
ES



Components of an ES



Forward Chaining Systems



Back
ward Chaining



Participants in Expert Systems Development and Use



Evolution of Expert Systems Software



Advantages of Expert Systems



Expert Systems Development Alternatives



Applications of Expert Systems an
d Artificial Intelligence



Study unit 7


23


7.5

ASSESSMENT QUESTIONS

Test Two


Study unit 7


24

Study unit 8


25


8

ATRIFICIAL NEURAL
NETWORKS






8.1

UNIT OUTCOMES

Learners will be able to:



Define an artificial neural network
with an aid of

a well labelled diagram



Compare and contrast an artificial neural network to a biological neuron



Draw and label both the artificial neural network and the biological neuron



Compare and contrast the back propagation to feed forward algorithm



Discuss and an
alyze the training methods



8.2

UNIT OVERVIEW

Artificial

Neural Networks is a field of artificial intelligence which is biologically inspired. It is
modelled after a human brain and it consist
s

of all the features of the biological neuron such
as synapse, de
ndrite, Axon, Nucleus, and soma. The neural Net models mathematically, all
these features of the biological neuron. Given the structure and functionality of the Artificial

Neural networks, the unit discusses the application of this technology in business a
nd in
industry. Its capabilities are explored in forecasting, prediction and in pattern recognition

(in
prognosis)
. The Artificial Neural Network algorithms are also explored.



Study unit 8


26


8.3

UNIT READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Appro
ach. Second Edition.
Prentice Hall. 2003.



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesley. 2005.



Hofstadter, DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial
Intelligence: A Guide to Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electrical Engineering and Computer Science,
University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660
Publisher:

Addison
-
Wesley



8.4

IMPORTANT UNIT CONCE
PTS



History of Artificial Neural Networks



Application of ANN



Strengths of a Neural Network



Advantages and Disadvantages of ANN



Status of Neural Networks



Neural networks Paradigm



Application of

neural network systems



The l
imitations

of
Neural Net
s



The Key Elem
ents of Neural Networks



Biological neuron



Layered Networks



SISO Single Hidden Layer Network



Training Data Set



Training Weights: Error Back
-
Propagation



Weight update formula



Back propagation

Algorithm



Error Back
-
Propagation (BP)

Study unit 8


27



Training methods



Feed forwar
d NNs





8.5

ASSESSMENT QUESTIONS



Test Two



Label the given biological neuron diagram


Study unit 8


28

Study unit 9


29



9

INTELLIGENT SYSTEMS






9.1

UNIT OUTCOMES

Learners will be able to:



Describe features of some intelligent systems



Identify
and discuss all the sub division of intelligent systems



Identify areas where intelligent systems are implement and how they aid humanity



Discuss in a logical way why some agents are referred to as general why some are
referred to as intelligent agents



9.2

U
NIT OVERVIEW

The unit introduces learners to the paradigm of intelligent systems. It further classifies the
intelligent systems to subdivisions such as intelligent agents, neural computing, fuzzy logic,
natural language processing, and expert systems. The
Intelligent agents can be classified as
software agents which
,

in turn can be broadly classified as general intelligent agent and
intelligent agents. The unit also explores examples of intelligent algorithms such as: Artificial
Neural Networks (ANN), Induc
tive Learning, Case
-
based Reasoning and Analogical
Reasoning, Genetic Algorithms, and Fuzzy Logic.



Study unit 9


30


9.3

UNIT READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Prentice Hall. 2003.



LUGER, GF, Artificial Intelligence

Structures and Strategies for Complex Problem
Solving. Fifth Edition. Addison Wesley. 2005.



Hofstadter, DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide to Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electrical Engineering and Computer Science,
University of Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660
Publisher:

Addison
-
Wesley



9.4

IMPORTANT UNIT CONCE
PTS



Properties of Intelligent systems



Intelligent Systems



Objectives of Intelli
gent systems



Examples of Intelligent Algorithms



Intelligent systems in business



Characteristics of intelligent systems



Software Agents



The “General
-

Intelligent Agent”



Intelligent Agent (IA)



Major Tasks Performed by IA



Applications of IA



Overview of Inte
lligent System Methodologies



Expert Systems



Overview of Expert Systems



Intelligent software agents (ISA)


Study unit 9


31


9.5

ASSESSMENT QUESTIONS



Test Three


Study unit 9


32

Study unit 10


33


10

GENETIC ALGORITHMS







10.1

UNIT OUTCOMES

Learners will be able to:



Define and discuss the characteristics of genetic algorithms



Compare and contrast genetic algorithms to natural selection



List and discuss stochastic operators



Analyze and implement genetic algorithms



Examine the evolutionary cycle of genetic algorithms a
nd contrast it to the natural
selection and reproduction cycles



10.2

UNIT OVERVIEW

The unit gives learners a general overview and appreciation of genetic algorithms. The basic
principles and reproductive processes implemented by genetic algorithms are discus
sed.
The fundamental design objective of genetic algorithms is reproduction
,

modelled after the
natural selection where the survival of the fittest is the rule of thumb. The genetic algorithm
consists of the following natural processes: reproduction, Comp
etition, selection, and
survival.



Study unit 10


34


10.3

UNIT READINGS



RUSSELL, S & NORVIG, P, Artificial Intelligence A Modern Approach. Second Edition.
Prentice Hall. 2003.



LUGER, GF, Artificial Intelligence Structures and Strategies for Complex Problem
Solving. Fifth Edit
ion. Addison Wesley. 2005.



Hofstadter, DR, Gödel, Esher, Bach: An Eternal Golden Braid. Vintage. 1989.



Artificial Intelligence: A Guide to Intelligent Systems, 2/E

Michael
Negnevitsky
,

School of Electrical Engineering and Computer Science,
University of
Tasmania

ISBN
-
10:

0321204662 ISBN
-
13:

9780321204660
Publisher:

Addison
-
Wesley



10.4

IMPORTANT UNIT CONCE
PTS



Overview of Genetic Algorithms



Classes of Search Techniques



Stochastic operators



Comparison of GA and Natural Selection



Simple Genetic Algorithm



The Evolutionary Cycle



Genetic Algorithm



Conceptual Algorithm



Reproduction



Reproduction Operators


10.5

ASSESSMENT QUESTIONS

Test three


summary of key areas of the module

Study unit 10


35

10.6

UNIT MODERATION

The first examiner is Mthulisi Velempini

and the second examiner is Prof Sam Lubbe. An
external examiner will be appointed by the department to review projects and the fin
al
examination.



Lecturer:

Mthulisi Velempini



Office: GAB A
1
32



Consultation Hours: Mondays 0745


1900 hours and Thursdays
0945


1245 hours



Phone: 018 389 2056



E
-
mail: mthulisi.velempini@nwu.ac.za