1.
IntelligentIntelligent
SystemsSystems
Unit Coordinator:Unit Coordinator:
SHAHIN ELAHI BOXSHAHIN ELAHI BOX
Email: shahinye@hotmail.comEmail: shahinye@hotmail.com

2.
22
Unit aimsUnit aims
 to be aware of the rationale of the artificial intelligenceto be aware of the rationale of the artificial intelligence
and soft computing paradigms with their advantagesand soft computing paradigms with their advantages
over traditional computingover traditional computing
 to gain an understanding of the theoretical foundationsto gain an understanding of the theoretical foundations
of various types of intelligent systems technologies to aof various types of intelligent systems technologies to a
level adequate for achieving objectives as stated belowlevel adequate for achieving objectives as stated below
 to develop the ability to evaluate intelligent systems,to develop the ability to evaluate intelligent systems,
and in particular, their suitability for specificand in particular, their suitability for specific
applicationsapplications
 to be able to manage the application of various toolsto be able to manage the application of various tools
available for developing intelligent systemsavailable for developing intelligent systems

3.
33
Unit delivery andUnit delivery and
learning structurelearning structure
 3 hours of lecture/workshop per week3 hours of lecture/workshop per week
 Lecture/WS time will be spent discussing the relevantLecture/WS time will be spent discussing the relevant
topic after an introduction by the lecturertopic after an introduction by the lecturer
 Topic lecture notes will be available early in the weekTopic lecture notes will be available early in the week
 Students should make use of the topic reading materialStudents should make use of the topic reading material
in advance for the topic to be coveredin advance for the topic to be covered
 Bringing up issues and questions for discussion areBringing up issues and questions for discussion are
encouraged to create an interactive learningencouraged to create an interactive learning
environment (this is assessed).environment (this is assessed).

4.
44
Resources and TextbooksResources and Textbooks
 Main text:Main text:
 Negnevitsky, M. Artificial Intelligence: A GuideNegnevitsky, M. Artificial Intelligence: A Guide
to Intelligent Systems, 2005. 2nd Edition.to Intelligent Systems, 2005. 2nd Edition.
 The main text to be supplemented by chapters/articlesThe main text to be supplemented by chapters/articles
from other books/journals/magazines as well as notesfrom other books/journals/magazines as well as notes
provided by the unit coordinator.provided by the unit coordinator.
 A list of recommended readings and other resourcesA list of recommended readings and other resources
will be provided for each topic.will be provided for each topic.
 Unit website:Unit website:
http://www.it.murdoch.edu.au/units/ICT619http://www.it.murdoch.edu.au/units/ICT619 willwill
enable access to unit reading materials and links toenable access to unit reading materials and links to
other resources.other resources.

5.
55
AssessmentAssessment
ACTIVITYACTIVITY DUEDUE WEIGHTWEIGHT
WorkshopWorkshop
participationparticipation
ContinuousContinuous 10%10%
ProjectProject Week 12Week 12 35%35%
Closed-bookClosed-book
ExamExam
Nov exams periodNov exams period 55%55%

6.
66
Topic scheduleTopic schedule
 Topic 1:Topic 1: Introduction to Intelligent Systems:Introduction to Intelligent Systems:
Tools, Techniques and ApplicationsTools, Techniques and Applications
 Topic 2:Topic 2: Rule-Based Expert SystemsRule-Based Expert Systems
 Topic 3:Topic 3: Fuzzy SystemsFuzzy Systems
 Topic 4:Topic 4: Neural ComputingNeural Computing
 Topic 5:Topic 5: Genetic AlgorithmsGenetic Algorithms
 Topic 6:Topic 6: Case-based ReasoningCase-based Reasoning
 Topic 7:Topic 7: Data MiningData Mining
 Topic 8:Topic 8: Intelligent Software AgentsIntelligent Software Agents
 Topic 9:Topic 9: Language TechnologyLanguage Technology

7.
77
Topic 1: Introduction toTopic 1: Introduction to
Intelligent SystemsIntelligent Systems
 What is an intelligent system?What is an intelligent system?
 Significance of intelligent systems in businessSignificance of intelligent systems in business
 Characteristics of intelligent systemsCharacteristics of intelligent systems
 The field of Artificial Intelligence (AI)The field of Artificial Intelligence (AI)
 The Soft Computing paradigmThe Soft Computing paradigm
 An Overview of Intelligent System MethodologiesAn Overview of Intelligent System Methodologies
 Expert SystemsExpert Systems
 Fuzzy SystemsFuzzy Systems
 Artificial Neural NetworksArtificial Neural Networks
 Genetic Algorithms (GA)Genetic Algorithms (GA)
 Case-based reasoning (CBR)Case-based reasoning (CBR)
 Data MiningData Mining
 Intelligent Software AgentsIntelligent Software Agents

8.
88
What is an intelligent system?What is an intelligent system?
 What is intelligence?What is intelligence?
 Hard to define unless you list characteristics eg,Hard to define unless you list characteristics eg,
 ReasoningReasoning
 LearningLearning
 AdaptivityAdaptivity
 A truly intelligent system adapts itself to deal withA truly intelligent system adapts itself to deal with
changes in problems (automatic learning)changes in problems (automatic learning)
 Few machines can do that at presentFew machines can do that at present
 Machine intelligenceMachine intelligence has a computer follow problemhas a computer follow problem
solving processes something like that in humanssolving processes something like that in humans
 Intelligent systemsIntelligent systems display machine-level intelligence,display machine-level intelligence,
reasoning, often learning, not necessarily self-adaptingreasoning, often learning, not necessarily self-adapting

9.
99
Intelligent systems in businessIntelligent systems in business
 Intelligent systems in business utilise one or moreIntelligent systems in business utilise one or more intelligenceintelligence
tools,tools, usuallyusually to aid decision makingto aid decision making
 Provides business intelligence toProvides business intelligence to
 Increase productivityIncrease productivity
 Gain competitive advantageGain competitive advantage
 Examples of business intelligence – information onExamples of business intelligence – information on
 Customer behaviour patternsCustomer behaviour patterns
 Market trendMarket trend
 Efficiency bottlenecksEfficiency bottlenecks
 Examples of successful intelligent systems applications inExamples of successful intelligent systems applications in
business:business:
 Customer service (Customer Relations Modelling)Customer service (Customer Relations Modelling)
 Scheduling (eg Mine Operations)Scheduling (eg Mine Operations)
 Data miningData mining
 Financial market predictionFinancial market prediction
 Quality controlQuality control

10.
1010
Intelligent systems in businessIntelligent systems in business
– some examples– some examples
 HNC (now Fair Isaac) software’s credit card fraud detector FalconHNC (now Fair Isaac) software’s credit card fraud detector Falcon
offers 30-70% improvement over existing methods (an example ofoffers 30-70% improvement over existing methods (an example of
a neural network).a neural network).
 MetLife insurance uses automated extraction of information fromMetLife insurance uses automated extraction of information from
applications in MITA (an example of language technology use)applications in MITA (an example of language technology use)
 Personalized, Internet-based TV listings (an intelligent agent)Personalized, Internet-based TV listings (an intelligent agent)
 Hyundai’s development apartment construction plans FASTrak-Hyundai’s development apartment construction plans FASTrak-
Apt (a Case Based Reasoning project)Apt (a Case Based Reasoning project)
 US Occupational Safety and Health Administration (OSHA usesUS Occupational Safety and Health Administration (OSHA uses
"expert advisors" to help identify fire and other safety hazards at"expert advisors" to help identify fire and other safety hazards at
work sites (an expert system).work sites (an expert system).
Source: http://www.newsfactor.com/perl/story/16430.htmlSource: http://www.newsfactor.com/perl/story/16430.html

11.
1111
Characteristics of intelligentCharacteristics of intelligent
systemssystems
 Possess one or more of these:Possess one or more of these:
 Capability to extract and store knowledgeCapability to extract and store knowledge
 Human like reasoning processHuman like reasoning process
 Learning from experience (or training)Learning from experience (or training)
 Dealing with imprecise expressions of factsDealing with imprecise expressions of facts
 Finding solutions through processes similar to natural evolutionFinding solutions through processes similar to natural evolution
 Recent trendRecent trend
 More sophisticated Interaction with the user throughMore sophisticated Interaction with the user through
 natural language understandingnatural language understanding
 speech recognition and synthesisspeech recognition and synthesis
 image analysisimage analysis
 Most current intelligent systems are based onMost current intelligent systems are based on
 rule based expert systemsrule based expert systems
 one or more of the methodologies belonging toone or more of the methodologies belonging to soft computingsoft computing

12.
1212
The field of Artificial Intelligence (AI)The field of Artificial Intelligence (AI)
 Primary goal:Primary goal:
 Development of software aimed at enabling machines to solveDevelopment of software aimed at enabling machines to solve
problems through human-like reasoningproblems through human-like reasoning
 Attempts to build systems based on a model of knowledgeAttempts to build systems based on a model of knowledge
representation and processing in the human mindrepresentation and processing in the human mind
 Encompasses study of the brain to understand its structure andEncompasses study of the brain to understand its structure and
functionsfunctions
 In existence as a discipline since 1956In existence as a discipline since 1956
 Failed to live up to initial expectations due toFailed to live up to initial expectations due to
 inadequate understanding of intelligence, brain functioninadequate understanding of intelligence, brain function
 complexity of problems to be solvedcomplexity of problems to be solved
 Expert systems – an AI success story of the 80sExpert systems – an AI success story of the 80s
 Case Based Reasoning systems - partial successCase Based Reasoning systems - partial success

13.
1313
The Soft Computing (SC) paradigmThe Soft Computing (SC) paradigm
 Also known asAlso known as Computational IntelligenceComputational Intelligence
 Unlike conventional computing, SC techniquesUnlike conventional computing, SC techniques
1.1. can be tolerant of imprecise, incomplete or corrupt input datacan be tolerant of imprecise, incomplete or corrupt input data
2.2. solve problems without explicit solution stepssolve problems without explicit solution steps
3.3. learn the solution through repeated observation andlearn the solution through repeated observation and
adaptationadaptation
4.4. can handle information expressed in vague linguistic termscan handle information expressed in vague linguistic terms
5.5. arrive at an acceptable solution through evolutionarrive at an acceptable solution through evolution

14.
1414
The Soft Computing (SC) paradigmThe Soft Computing (SC) paradigm
(cont’d)(cont’d)
 The first four characteristics are common inThe first four characteristics are common in
problem solving by individual humansproblem solving by individual humans
 The fifth characteristic (evolution) is common inThe fifth characteristic (evolution) is common in
naturenature
 The predominant SC methodologies found inThe predominant SC methodologies found in
current intelligent systems are:current intelligent systems are:
 Artificial Neural Networks (ANN)Artificial Neural Networks (ANN)
 Fuzzy SystemsFuzzy Systems
 Genetic Algorithms (GA)Genetic Algorithms (GA)

15.
1515
Overview of Intelligent SystemOverview of Intelligent System
MethodologiesMethodologies
- Expert Systems (ES)- Expert Systems (ES)
 Designed to solve problems in aDesigned to solve problems in a specific domainspecific domain,,
 eg, an ES to assist foreign currency traderseg, an ES to assist foreign currency traders
 Built byBuilt by
 interrogating domain expertsinterrogating domain experts
 storing acquired knowledge in a form suitable for solvingstoring acquired knowledge in a form suitable for solving
problems, using simple reasoningproblems, using simple reasoning
 Used byUsed by
 Querying the user for problem-specific informationQuerying the user for problem-specific information
 Using the information to draw inferences from the knowledgeUsing the information to draw inferences from the knowledge
basebase
 Supplies answers or suggested ways to collect further inputsSupplies answers or suggested ways to collect further inputs

16.
1616
Overview of Expert SystemsOverview of Expert Systems
(cont’d)(cont’d)
 Usual form of the expert system knowledgeUsual form of the expert system knowledge
base is a collection of IF … THEN … rulesbase is a collection of IF … THEN … rules
 Note: not IF statements in procedural codeNote: not IF statements in procedural code
 Some areas of ES application:Some areas of ES application:
 banking and finance (credit assessment, projectbanking and finance (credit assessment, project
viability)viability)
 maintenance (diagnosis of machine faults)maintenance (diagnosis of machine faults)
 retail (suggest optimal purchasing pattern)retail (suggest optimal purchasing pattern)
 emergency services (equipment configuration)emergency services (equipment configuration)
 law (application of law in complex scenarios)law (application of law in complex scenarios)

17.
1717
Artificial Neural Networks (ANN)Artificial Neural Networks (ANN)
 Human brain consists of 100 billion densely interconnected simpleHuman brain consists of 100 billion densely interconnected simple
processing elements known as neuronsprocessing elements known as neurons
 ANNs are based on a simplified model of the neurons and theirANNs are based on a simplified model of the neurons and their
operationoperation
 ANNs usually learn from experience – repeated presentation ofANNs usually learn from experience – repeated presentation of
example problems with their corresponding solutionsexample problems with their corresponding solutions
 After learning the ANN is able to solve problems, even withAfter learning the ANN is able to solve problems, even with
newish inputnewish input
 The learning phase may or may not involve human interventionThe learning phase may or may not involve human intervention
(supervised vs unsupervised learning)(supervised vs unsupervised learning)
 The problem solving 'model' developed remains implicit andThe problem solving 'model' developed remains implicit and
unknown to the userunknown to the user
 Particularly suitable for problems not prone to algorithmicParticularly suitable for problems not prone to algorithmic
solutions, eg, pattern recognition, decision supportsolutions, eg, pattern recognition, decision support

18.
1818
Artificial Neural Networks (cont’d)Artificial Neural Networks (cont’d)
 Different models of ANNs depending onDifferent models of ANNs depending on
 ArchitectureArchitecture
 learning methodlearning method
 other operational characteristics (eg type of activation function)other operational characteristics (eg type of activation function)
 Good at pattern recognition and classification problemsGood at pattern recognition and classification problems
 Major strength - ability to handle previously unseen, incomplete orMajor strength - ability to handle previously unseen, incomplete or
corrupted datacorrupted data
 Some application examples:Some application examples:
- explosive detection at airports- explosive detection at airports
- face recognition- face recognition
- financial risk assessment- financial risk assessment
- optimisation and scheduling- optimisation and scheduling

19.
1919
Genetic Algorithms (GA)Genetic Algorithms (GA)
 Belongs to a broader field known asBelongs to a broader field known as evolutionary computationevolutionary computation
 Solution obtained by evolving solutions through a processSolution obtained by evolving solutions through a process
consisting ofconsisting of
 survival of the fittestsurvival of the fittest
 crossbreeding, andcrossbreeding, and
 mutationmutation
 A population of candidate solutions is initialised (theA population of candidate solutions is initialised (the
chromosomes)chromosomes)
 New generations of solutions are produced beginning with theNew generations of solutions are produced beginning with the
intial population, using specific genetic operations: selection,intial population, using specific genetic operations: selection,
crossover and mutationcrossover and mutation

20.
2020
Genetic Algorithms (cont’d)Genetic Algorithms (cont’d)
 Next generation of solutions produced from the current populationNext generation of solutions produced from the current population
usingusing
 crossover (splicing and joining peices of the solution from parents) andcrossover (splicing and joining peices of the solution from parents) and
 mutation (random change in the parameters defining the solution)mutation (random change in the parameters defining the solution)
 TheThe fitnessfitness of newly evolved solution evaluated using a fitnessof newly evolved solution evaluated using a fitness
functionfunction
 The steps of solution generation and evaluation continue until anThe steps of solution generation and evaluation continue until an
acceptable solution is foundacceptable solution is found
 GAs have been used inGAs have been used in
 portfolio optimisationportfolio optimisation
 bankruptcy predictionbankruptcy prediction
 financial forecastingfinancial forecasting
 design of jet enginesdesign of jet engines
 schedulingscheduling

21.
2121
Fuzzy SystemsFuzzy Systems
 Traditional logic is two-valued – any proposition isTraditional logic is two-valued – any proposition is
either true or falseeither true or false
 Problem solving in real-life must deal with partially trueProblem solving in real-life must deal with partially true
or partially false propositionsor partially false propositions
 Imposing precision may be difficult and lead to lessImposing precision may be difficult and lead to less
than optimal solutionsthan optimal solutions
 Fuzzy systems handle imprecise information byFuzzy systems handle imprecise information by
assigning degrees of truth - using fuzzy logicassigning degrees of truth - using fuzzy logic

22.
2222
Fuzzy Systems (cont’d)Fuzzy Systems (cont’d)
 FL allow us to express knowledge in vague linguisticFL allow us to express knowledge in vague linguistic
termsterms
 Flexibility and power of fuzzy systems now wellFlexibility and power of fuzzy systems now well
recognised (eg simplification of rules in control systemsrecognised (eg simplification of rules in control systems
where imprecision is found)where imprecision is found)
 Some applications of fuzzy systems:Some applications of fuzzy systems:
 Control of manufacturing processesControl of manufacturing processes
 appliances such as air conditioners, washing machines andappliances such as air conditioners, washing machines and
video camerasvideo cameras
 Used in combination with other intelligent systemUsed in combination with other intelligent system
methodologies to develop hybrid fuzzy-expert, neuro-fuzzy,methodologies to develop hybrid fuzzy-expert, neuro-fuzzy,
or fuzzy-GA systemsor fuzzy-GA systems

23.
2323
Case-based reasoning (CBR)Case-based reasoning (CBR)
 CBR systems solve problems by making use of knowledge aboutCBR systems solve problems by making use of knowledge about
similar problems encountered in the pastsimilar problems encountered in the past
 The knowledge used in the past is built up as a case-baseThe knowledge used in the past is built up as a case-base
 CBR systems search the case-base for cases with attributesCBR systems search the case-base for cases with attributes
similar to given problemsimilar to given problem
 A solution created by synthesizing similar cases, and adjusting toA solution created by synthesizing similar cases, and adjusting to
cater for differences between given problem and similar casescater for differences between given problem and similar cases
 Difficult to do well in practice, but very powerful if you can do itDifficult to do well in practice, but very powerful if you can do it

24.
2424
Case-based reasoning (cont’d)Case-based reasoning (cont’d)
 CBR systems can improve over time by learning fromCBR systems can improve over time by learning from
mistakes made with past problemsmistakes made with past problems
 Application examples:Application examples:
 Utilisation of shop floor expertise in aircraft repairsUtilisation of shop floor expertise in aircraft repairs
 Legal reasoningLegal reasoning
 Dispute mediationDispute mediation
 Data miningData mining
 Fault diagnosisFault diagnosis
 SchedulingScheduling

25.
2525
Data miningData mining
 The process of exploring and analysing data forThe process of exploring and analysing data for
discovering new and useful informationdiscovering new and useful information
 Huge volumes of mostly point-of-sale (POS) data areHuge volumes of mostly point-of-sale (POS) data are
generated or captured electronically every day, eg,generated or captured electronically every day, eg,
 data generated by bar code scannersdata generated by bar code scanners
 customer call detail databasescustomer call detail databases
 web log files in e-commerce etc.web log files in e-commerce etc.
 Organizations are ending up with huge amounts ofOrganizations are ending up with huge amounts of
mostly day-to-day transaction datamostly day-to-day transaction data

26.
2626
Data mining (cont’d)Data mining (cont’d)
 It is possible to extract useful information on market and customerIt is possible to extract useful information on market and customer
behaviour by “mining" the databehaviour by “mining" the data
 Note: This goes far beyond simple statistical analysis of numericalNote: This goes far beyond simple statistical analysis of numerical
data, to classification and analysis of non-numerical datadata, to classification and analysis of non-numerical data
 Such information mightSuch information might
 reveal important underlying trends and associations in marketreveal important underlying trends and associations in market
behaviour, andbehaviour, and
 help gain competitive advantage by improving marketinghelp gain competitive advantage by improving marketing
effectivenesseffectiveness
 Techniques such as artificial neural networks and decision treesTechniques such as artificial neural networks and decision trees
have made it possible to perform data mining involving largehave made it possible to perform data mining involving large
volumes of data (from "data warehouses").volumes of data (from "data warehouses").
 Growing interest in applying data mining in areas such directGrowing interest in applying data mining in areas such direct
target marketing campaigns, fraud detection, and development oftarget marketing campaigns, fraud detection, and development of
models to aid in financial predictions, antiterrorism systemsmodels to aid in financial predictions, antiterrorism systems

27.
2727
Intelligent software agentsIntelligent software agents (ISA)(ISA)
 ISAs are computer programs that provide active assistance toISAs are computer programs that provide active assistance to
information system usersinformation system users
 Help users cope with information overloadHelp users cope with information overload
 Act in many ways like a personal assistant to the user byAct in many ways like a personal assistant to the user by
attempting to adapt to the specific needs of the userattempting to adapt to the specific needs of the user
 Capable of learning from the user as well as other intelligentCapable of learning from the user as well as other intelligent
software agentssoftware agents
 Application examples:Application examples:
 News and Email Collection,News and Email Collection,
Filtering and ManagementFiltering and Management
 Online ShoppingOnline Shopping
 Event NotificationEvent Notification
 Personal schedulingPersonal scheduling
 Online help desks, interactive charactersOnline help desks, interactive characters
 Rapid Response ImplementationRapid Response Implementation

28.
2828
Language Technology (LT)Language Technology (LT)
 ““[The] application of knowledge about human language in computer-[The] application of knowledge about human language in computer-
based solutions” (Dale 2004)based solutions” (Dale 2004)
 Communication between people and computers is an importantCommunication between people and computers is an important
aspect of any intelligent information systemaspect of any intelligent information system
 Applications of LT:Applications of LT:
 Natural Language Processing (NLP)Natural Language Processing (NLP)
 Knowledge RepresentationKnowledge Representation
 Speech recognitionSpeech recognition
 Optical character recognition (OCR)Optical character recognition (OCR)
 Handwriting recognitionHandwriting recognition
 Machine translationMachine translation
 Text summarisationText summarisation
 Speech synthesisSpeech synthesis
 A LT-based system can be the front-end ofA LT-based system can be the front-end of
information systems themselves based oninformation systems themselves based on
other intelligence toolsother intelligence tools
Hi, I am Cybelle.
What is your name?

29.
2929
For Next WeekFor Next Week
Get hold of the textbookGet hold of the textbook
 Visit the library and find the section onVisit the library and find the section on
artificial intelligence, browse some titlesartificial intelligence, browse some titles
 Get onto the unit website, download andGet onto the unit website, download and
read papers concerning Expert Systemsread papers concerning Expert Systems
 We will study the theory and practiceWe will study the theory and practice
developing a simple expert systemdeveloping a simple expert system