Department of Computer Science, - :: Prof. Dr. S.M. Aqil ... PROFESSOR DR. S.M. AQIL BURNEY Department of Computer Science at UBIT ... “Artificial intelligence is the study of how

Department of Computer Science, Department of Computer Science,

(UBIT)(UBIT)

University of KarachiUniversity of Karachi

Artificial Intelligence: Artificial Intelligence: Logic Systems for Logic Systems for

ReasoningReasoning

(First National Conference on (First National Conference on Mathematical Sciences)Mathematical Sciences)

2222––MarchMarch––2010 2010

MERITORIOUS PROFESSOR DR. S.M. AQIL BURNEYDepartment of Computer Science at UBIT

University of KarachiEmail: [email protected]

www.drburney.net

2222––MarchMarch––2010 2010

•• Hard Approach : Precise Truth values Hard Approach : Precise Truth values available for modeling and computingavailable for modeling and computing

Modeling and ComputingModeling and Computing

•• Soft Approach : Imprecise Truth valuesSoft Approach : Imprecise Truth values

What is Intelligence?What is Intelligence?

�� IntelligenceIntelligence is the ability to understand is the ability to understand and learn things. and learn things.

�� IntelligenceIntelligence is the ability to think and is the ability to think and �� IntelligenceIntelligence is the ability to think and is the ability to think and understand instead of doing things by understand instead of doing things by instinct or automatically.instinct or automatically.

((Essential English DictionaryEssential English Dictionary, Collins, London, , Collins, London, 1990)1990)

Definition of Artificial Definition of Artificial

IntelligenceIntelligence�� “Artificial intelligence is the study of how to make “Artificial intelligence is the study of how to make

computers do things at which, at the moment, people are computers do things at which, at the moment, people are better”.better”.

�� “ Artificial intelligence is the branch of Computer Science “ Artificial intelligence is the branch of Computer Science that deals with ways of representing knowledge using that deals with ways of representing knowledge using symbols rather than numbers and with rules of thumb, or symbols rather than numbers and with rules of thumb, or heuristic, methods for processing information”. heuristic, methods for processing information”. heuristic, methods for processing information”. heuristic, methods for processing information”. Bruce G. Buchanan, Encyclopedia BritannicaBruce G. Buchanan, Encyclopedia Britannica

�� “ In simplified terms, artificial intelligence works with “ In simplified terms, artificial intelligence works with pattern matching methods which attempt to describe pattern matching methods which attempt to describe objects, events, or processes in terms of their qualitative objects, events, or processes in terms of their qualitative features and logical and computational relationships”. features and logical and computational relationships”. Brattle Research Corporation, Artificial Intelligence and Fifth Brattle Research Corporation, Artificial Intelligence and Fifth Generation Computer TechnologiesGeneration Computer Technologies

Goals of AIGoals of AI

�� Scientific ,Business and engineeringScientific ,Business and engineering•• Understanding of computational Understanding of computational mechanisms needed for intelligent mechanisms needed for intelligent behaviorbehavior

•• Intelligent connection of perception Intelligent connection of perception and actionand actionand actionand action

•• Replicate human intelligenceReplicate human intelligence•• Solve knowledgeSolve knowledge--intensive tasksintensive tasks•• Enhance humanEnhance human--human, humanhuman, human--computer and computercomputer and computer--computer computer interaction/communicationinteraction/communication

What Computers can do better than People

•Numerical Computation

•Information Storage•Information Storage

•Repetitive Operations

What people can do better than Computers

•Understand Information

•Common Sense

•Inference

•Pattern Matching

•Intuition

What problems the people are What problems the people are

facing in ICT age?facing in ICT age?�� A professor of psychology at Harvard University A professor of psychology at Harvard University

warned that by the year 2000 the limit of man’s mind warned that by the year 2000 the limit of man’s mind to absorb information may be reachedto absorb information may be reached

�� Most recently very complex and technologically Most recently very complex and technologically oriented society has created a situation in which more oriented society has created a situation in which more people an organizations have become concerned with people an organizations have become concerned with handling information and fewer with handling materialhandling information and fewer with handling materialpeople an organizations have become concerned with people an organizations have become concerned with handling information and fewer with handling materialhandling information and fewer with handling material

�� Information is an essential element in decision makingInformation is an essential element in decision making

�� One of the major problems in design of modern One of the major problems in design of modern information systems is automatic pattern recognition. information systems is automatic pattern recognition. Which forms the dominant part of Computer Science Which forms the dominant part of Computer Science and Information Technology.and Information Technology.

�� J. T. Tou, R.C. Gonzalez; Pattern recognition Principles J. T. Tou, R.C. Gonzalez; Pattern recognition Principles Addison Wesley Publishing Company, LondonAddison Wesley Publishing Company, London

Fields of AI

•Expert Systems

•Natural Language Processing

•Speech Recognition

•Computer Vision•Computer Vision

•Robotics

•Intelligent Computer Assisted Instruction

•Automatic Programming

•Planning and Decision Support

Artificial Life

Modeling Hu

man

Performance

ArtificialIntelligence

Machine

Learning &

Robotics

AI & Phil

osophy

Neural

Networks Parallel

Distribute processing

Automated

Reasoning

Natural

Language

Processing

Expert Systems Distributed AI

Playing Game

�� Computer ScienceComputer Science

�� CriminologyCriminology�� Cognitive ScienceCognitive Science

Kinesiology & Health SciencesKinesiology & Health Sciences

�� Information TechnologyInformation Technology

�� Mathematics Mathematics

�� Religious StudiesReligious Studies

�� PsychologyPsychology

�� Professional WritingProfessional Writing

�� WomenWomen’’s Studiess Studies

�� Environmental ManagementEnvironmental Management

�� MusicMusic

MusicMusic

�� Film & VideoFilm & Video

�� BiologyBiology �� LatinLatin

�� Creative WritingCreative Writing

�� Kinesiology & Health SciencesKinesiology & Health Sciences

�� Political SciencePolitical Science

�� Communication StudiesCommunication Studies

�� Theatre Theatre

�� MusicMusic

�� Translation Translation

�� Physics & Astronomy Physics & Astronomy

�� Law & SocietyLaw & Society

�� International StudiesInternational Studies

�� EconomicsEconomics �� European StudiesEuropean Studies

�� Health & SocietyHealth & Society

�� Classical StudiesClassical Studies

�� Computer Science/ICTComputer Science/ICT

�� CriminologyCriminology�� Cognitive ScienceCognitive Science

Kinesiology & Health SciencesKinesiology & Health Sciences

�� Information TechnologyInformation Technology

�� Mathematics Mathematics

�� Religious StudiesReligious Studies

�� PsychologyPsychology

�� Professional WritingProfessional Writing

�� WomenWomen’’s Studiess Studies

�� Environmental ManagementEnvironmental Management

�� MusicMusic

MusicMusic

�� Film & VideoFilm & Video

�� BiologyBiology �� LatinLatin

�� Creative WritingCreative Writing

�� Kinesiology & Health SciencesKinesiology & Health Sciences

�� Political SciencePolitical Science

�� Communication StudiesCommunication Studies

�� Theatre Theatre

�� MusicMusic

�� Translation Translation

�� Physics & Astronomy Physics & Astronomy

�� Law & SocietyLaw & Society

�� International StudiesInternational Studies

�� EconomicsEconomics �� European StudiesEuropean Studies

�� Health & SocietyHealth & Society

�� Classical StudiesClassical Studies

Life and IntelligenceLife and Intelligence

�� Self organization and perpetuation Self organization and perpetuation leads to Intelligence and Intelligence leads to Intelligence and Intelligence created in noncreated in non--biological Systems is biological Systems is Artificial Intelligence.Artificial Intelligence.Artificial Intelligence.Artificial Intelligence.

Modeling Human Performance

Human intelligence is reference point in AI. The

understanding of human intelligence is essential and

open scientific challenge.

• AI engineers usually not committed to • AI engineers usually not committed to

“making their programs act like humans.

•AI engineer seldom ignore the humans solution.

The discipline that uses AI tools to explore the nature

of human intelligence is called Cognitive Science.

Automated Reasoning

• Logic lends itself to automation.

• A variety of problems can be attacked by

representing the problem description and representing the problem description and

relevant background information as logical

axioms and treating problem instances as

theorems to be proved.

Logic and ReasoningLogic and Reasoning

Logical Reasoning

Reasoning Subjective

Probabilistic Reasoning

Bayesian Networks

�� Using given knowledge and truth Using given knowledge and truth value help us to solve, understand value help us to solve, understand real life problems.real life problems.

EXAMPLEEXAMPLE

•• p: All mathematicians wear glassesp: All mathematicians wear glasses

•• q: Anyone who wears glasses is an q: Anyone who wears glasses is an algebraistalgebraist

•• r: All mathematicians are algebraistr: All mathematicians are algebraist•• r: All mathematicians are algebraistr: All mathematicians are algebraist

pp∧∧q q →→ r r ≡≡ ( ( ∼∼( p( p∧∧q) q) ∨∨ r)r)

TRUTH TABLE

Truth Table for the formulae built with the Logical Operators

p q r pΛΛΛΛq ~(pΛΛΛΛq) ~(pΛΛΛΛq)Vr

T T T T F T

T T F T F F

T F T F T TT F T F T T

T F F F T T

F T T F T T

F T F F T T

F F T F T T

F F F F T T

�� If r is the conclusion, and we know If r is the conclusion, and we know that p and q are true simultaneously that p and q are true simultaneously then r is valid statement.then r is valid statement.

�� In real life, the statements are true In real life, the statements are true or false, here statement means an or false, here statement means an or false, here statement means an or false, here statement means an atomic statement, thus statements atomic statement, thus statements may be simple (atomic) or may be simple (atomic) or component. If p, q and r are component. If p, q and r are independent statements, then we independent statements, then we

need to prove: need to prove: pp∧∧q q →→ r r

CommitmentCommitment

Ontological CommitmentOntological Commitment: : PPTPPT

What exists in the world: Language of What exists in the world: Language of reasoning (Formal).reasoning (Formal).

Epistemological CommitmentEpistemological CommitmentEpistemological CommitmentEpistemological Commitment

What an intelligent entity believes What an intelligent entity believes about the fact.about the fact.

Believe System: True, False, Believe System: True, False, Unknown, degree of believe, degree Unknown, degree of believe, degree believe with ranks (known values)believe with ranks (known values)

NUMBERS ARE RESPECTEDNUMBERS ARE RESPECTED——WORDS ARE NOTWORDS ARE NOT

�� “In science and engineering there is a deep“In science and engineering there is a deep--seated tradition of according much more seated tradition of according much more respect to numbers than to words. The respect to numbers than to words. The essence of this tradition was stated essence of this tradition was stated succinctly by Lord Kelvin in 1883.”succinctly by Lord Kelvin in 1883.”succinctly by Lord Kelvin in 1883.”succinctly by Lord Kelvin in 1883.”

Zadeh Zadeh -- 20042004

�� But now imprecise knowledge But now imprecise knowledge in words is also of immense in words is also of immense importance in developing importance in developing logical systems.logical systems.

FormalFormal

LanguageLanguage

OntologyOntology

What exists)What exists)

EpistemologyEpistemology

PropositionPropositional Logical Logic

factsfacts True/False True/False /Unknown/Unknown

Predicate Predicate LogicLogic

Facts, objects, Facts, objects, relationsrelations

True/False True/False /Unknown/UnknownLogicLogic relationsrelations /Unknown/Unknown

Probability Probability TheoryTheory

Facts with Facts with changechange

Degree of Degree of believe on [0,1]believe on [0,1]

Temporal Temporal LogicLogic

Facts, objects, Facts, objects, relation and relation and timetime

True/False True/False /Unknown/Unknown

Fuzzy LogicFuzzy Logic(Lectures of Dr. (Lectures of Dr. Burney)Burney)

Facts with Facts with degree of degree of believebelieve

Known Known interval interval valuevalue

ANNANN--FLFL

(NFL (NFL Example)Example)

Facts with Facts with degree of degree of believe with believe with

Known Known interval interval values with values with Example)Example) believe with believe with

learninglearningvalues with values with improvement improvement in believein believe

Spatial LogicSpatial Logic Facts, Facts, objects, objects, relation, time relation, time & Space& Space

True/False True/False /Unknown/Unknown

Application of Fuzzy LogicApplication of Fuzzy Logic

�� An Example of Traffic Light Controller An Example of Traffic Light Controller SystemSystem

Evolution of NeuroEvolution of Neuro--Fuzzy LogicFuzzy Logic

Neural

Approximate Reasoning

Fuzzy Logic

Functional Approximation/

Randomized Search

Neuro-FuzzySystems

NeuralNetworks

Fuzzy Logic

Production systems

A production system provides pattern-solving process

and consists of a set of

• Production rule• Production rule

• A working Memory

• A recognize-act control cycle

Example of a production system

Problem statement:

Sort (re-arrange the alphabets in) a string

composed of only three alphabets a, b and c. for

Example the string cbaca should be sorted to

give out aabcc.give out aabcc.

Solution

To understand how this can be solved consider

the following three points

1. Production rules

To solve this problem we need production rules.

A Production rule is a structure used to

represent a variety of knowledge types. Such

rules are often described in an if … thenrules are often described in an if … thenfromat.

For this problem the Production set(i.e. set of rules to solve the problem)is given below

.ba → ab

.ca → ac.ca → ac

.cb → bc

Rules say that whenever the pattern“ba” is found replace it with “ab”, andso on.

The recognize act cycle

The current state of the problem solving

process is maintained as a set of

patterns in working memory. Working

memory is initialized with the beginning

problem description. Production rules areproblem description. Production rules are

then applied to current state of the

pattern. All the rules which match the

current pattern are noted and then

conflict resolution is performed. This

process is repeated until the set of

conflicting rules is null

Conflict resolution

The patterns in a working memory are matched against the conditions of the production rules; this produces a subset of the productions, called conflict set. Whose conditions match the patterns in working memory. The productions in the conflict set are said to be enabled. One of the are said to be enabled. One of the productions in the conflict set is then selected and a production rule is fired. That is the action of the rule is performed. Conflict resolution strategies may be as simple as selecting the first rule or may also be complex.

For this problem we choose “apply

rule with smallest rule number” asrule with smallest rule number” as

conflict resolution strategy.

Trace of the above production system.

Iteration # Working memory Conflict set Rule fired

0 “cbaca” 1, 2, 3 1

1 “cabca” 2 2

2 “acbca” 3, 2 22 “acbca” 3, 2 2

3 “acbac” 1, 3 1

4 “acabc” 2 2

5 “aacbc” 3 3

6 “aabcc” Null Halt

AI and Philosophy:

As human intelligence is reference As human intelligence is reference point to AI and HI evolves in point to AI and HI evolves in philosophical, mathematical, social philosophical, mathematical, social environment and have roots in environment and have roots in environment and have roots in environment and have roots in societies.societies.

ContributesContributes

HI AIHI AI

A design is made concrete in a program and the

program is run as experiment, then redesign,

rerun and testing and validity is done to conform

over representations and algorithms are sufficient

model of required intelligent behavior.

Software Development ModelsTuring Test

The necessary and sufficient condition for a

physical system to exhibit intelligence is that it

should be physical symbol system.

The Turing Test:

The Turing Test has been called the pinnacle of AI. In

the simplest case, a human judge sits in front of a

terminal. At the other end of the communication link

sits another human or a computer running AI

software. After an on-line chat, the judge decidessoftware. After an on-line chat, the judge decides

whether he or she has been communicating with a

computer or a human. If the judge can’t guess

correctly 70% of the time, the program passes the

Turing Test.

THE INTERROGATOR

Human Response

Machine Response

TURING TEST

Expert Systems:

An expert system is knowledge based program

that provides “expert quality solutions to problems

in a specific domain”.

Converting the knowledge of an expert in to a Converting the knowledge of an expert in to a

software usually give rise to an expert system--an

intelligent entity replaces an expert

In Simple

Human Experts in any field are frequently in great

Computer programs that embody human expertise

Human Experts in any field are frequently in great

demand and are, therefore, usually in short supply.

Limitation: Unfortunately in most fields there are

more problems than experts.

Components of an Expert System (SE)

Knowledge BaseKnowledge Base

Inference Engine

User

Interface

User

Communication

Knowledge Base

The component of ES that contains the systems

knowledge is called its knowledge base as meree

emulation of the human reasoning process is not

sufficient.

Declarative Knowledge (object Oriented)

facts about objects, events and situations

KB

Procedural Knowledge

(Information about possible courses of actions)

Knowledge Representation Techniques Knowledge Representation Techniques

Logic: For declarative Knowledge (DK)

Semantic Network: Graphical representation of DK

Production Systems: Procedural Knowledge is represented as production rules

about actions to be taken.

Rule-based expert Systems

Rule-based Systems

Expert systems development tools are needed to avoid expanses and delays and to improve efficiency of the development process.

ART TM (Automated Reasoning Tool TM )

useful for commercial expert systems for useful for commercial expert systems for

Resource scheduling, Manufacturing, Planning

Financial planning, command & control

• HES (Human Edge Software) Product

Expert Ease TM

• KEE TM (Knowledge Engineering Environment)

• KES (Knowledge Engineering SYSTEM)

• Expert Rule etc.

Natural Language Processing (NLP)

NLP Language that is typed, printed or displayed

rather than being spoken

Speech Recognition (SR)

SR Getting Computers to understand spoken

language is the focus of related AI technologies called

Speech Recognition.

Neural Network

Computing or information processing has been dominated

by the concept of “Programmed Computing” in which

i) Algorithms are designed

ii) Algorithms are implemented using currently

dominant architecture.

Computing in biological systems is much different from Computing in biological systems is much different from

above mentioned paradigm in that

i) The computation is massively distributed and parallel

ii) Learning replaces a prior program development

Artificial neural networks (ANN) are developed using

paradigm of computing in biological system such as

computation in brain.

�� Primary motor: voluntary movementPrimary motor: voluntary movement�� Primary somatosensory: tactile, pain, Primary somatosensory: tactile, pain,

pressure, position, temp., mvt.pressure, position, temp., mvt.�� Motor association: coordination of Motor association: coordination of

complex movementscomplex movements�� Sensory association: processing of Sensory association: processing of

multisensorial informationmultisensorial information�� Prefrontal: planning, emotion, judgementPrefrontal: planning, emotion, judgement�� Speech center (Broca’s area): speech Speech center (Broca’s area): speech

production and articulationproduction and articulation

Major Functional Areas

production and articulationproduction and articulation�� Wernicke’s area: comprehension of Wernicke’s area: comprehension of

speechspeech�� Auditory: hearingAuditory: hearing�� Auditory association: complex auditory Auditory association: complex auditory

processingprocessing�� Visual: lowVisual: low--level visionlevel vision�� Visual association: higherVisual association: higher--level visionlevel vision

Modeling of Brain Modeling of Brain

Functions: Functions:

InterconnectInterconnect

Structure of a NeuronStructure of a Neuron

Electron Micrograph of a Real Electron Micrograph of a Real

NeuronNeuron

Remember? Neurons & Synapses Remember? Neurons & Synapses

Key terms:Key terms:

�� AxonAxon

�� DendritesDendrites

�� SynapsesSynapses

�� Soma Soma (cell (cell (cell (cell body)body)

Engineering Applications: Engineering Applications: ClassificationClassification

•• Machinery defect diagnosis Machinery defect diagnosis

•• Signal processing Signal processing

•• Character recognition Character recognition

•• Process supervision Process supervision

•• Process fault analysis Process fault analysis •• Process fault analysis Process fault analysis

•• Speech recognition Speech recognition

•• Machine vision Machine vision

•• Speech recognition Speech recognition

•• Radar signal classificationRadar signal classification

Engineering Applications: Engineering Applications: ModellingModelling

•• Transducer linearization Transducer linearization

•• Color discrimination Color discrimination

•• Robot control and navigation Robot control and navigation

•• Process control Process control

•• Aircraft landing control Aircraft landing control •• Aircraft landing control Aircraft landing control

•• Car active suspension control Car active suspension control

•• Printed Circuit auto routing Printed Circuit auto routing

•• Integrated circuit layout Integrated circuit layout

•• Image compression Image compression

A MachineA Machine--gun Equipped Robotic gun Equipped Robotic Sentry.Sentry.

�� The sentry robot is equipped The sentry robot is equipped with two cameras, one for with two cameras, one for dayday--time and one for infrared time and one for infrared night vision, zooming night vision, zooming capabilities, a speaker for capabilities, a speaker for notifying the intruder of notifying the intruder of inpending death,.inpending death,.

�� Sophisticated pattern Sophisticated pattern recognition to detect the recognition to detect the

�� Sophisticated pattern Sophisticated pattern recognition to detect the recognition to detect the difference between humans difference between humans and trees, and a 5.5mm and trees, and a 5.5mm machinemachine--gun.gun.

�� http://www.gorobotics.net/Thhttp://www.gorobotics.net/Thee--News/Military/SouthNews/Military/South--KoreaKorea--DevelopsDevelops--Machine%11GunMachine%11Gun--SentrySentry--Robot/Robot/

Example: face recognitionExample: face recognition

�� Here using the 2Here using the 2--stage approach:stage approach:

TrainingTraining

�� http://www.nhttp://www.neci.nec.com/eci.nec.com/homepages/lhomepages/lawrence/papawrence/papers/faceers/face--ers/faceers/face--tr96/latex.httr96/latex.htmlml

AI AI –– Few AchievementsFew Achievements

�� Robots make cars in all advanced countries.Robots make cars in all advanced countries.�� Reasonable machine translation is available for a Reasonable machine translation is available for a

large range of foreign web pages.large range of foreign web pages.�� Computers land 200 ton jumbo jets unaided Computers land 200 ton jumbo jets unaided

every few minutes.every few minutes.�� Search systems like Google are not perfect but Search systems like Google are not perfect but �� Search systems like Google are not perfect but Search systems like Google are not perfect but

provide very effective information retrievalprovide very effective information retrieval�� Robots cut slots for hip joints better than Robots cut slots for hip joints better than

surgeons.surgeons.�� Deep blue beat Kasparov in 1997 and the current Deep blue beat Kasparov in 1997 and the current

world Go champion is a computer.world Go champion is a computer.�� Medical expert systems can outperform doctors in Medical expert systems can outperform doctors in

many areas of diagnosis.many areas of diagnosis.

Experts of Artificial Intelligence?Experts of Artificial Intelligence?

�� One of major divisions in AI (and you can see it One of major divisions in AI (and you can see it in the definitions on the previous slide) is in the definitions on the previous slide) is betweenbetween

�� Those who think AI is the only serious way of Those who think AI is the only serious way of finding finding out how out how wewe work (since opening heads does not work (since opening heads does not yet yet out how out how wewe work (since opening heads does not work (since opening heads does not yet yet give much insight into this) andgive much insight into this) and

�� Those who want computers to do very smart Those who want computers to do very smart things, things, independentlyindependently of how we work.of how we work.

�� This is the important distinction betweenThis is the important distinction betweenCognitive ScientistsCognitive Scientists vs. vs. EngineersEngineers..

AI AI –– A CommentA Comment

�� Despite all these achievements, one of the Despite all these achievements, one of the major philosophers of Cognitive Science major philosophers of Cognitive Science wrote recently:wrote recently:

“… the failure of artificial intelligence to “… the failure of artificial intelligence to produce successful simulation of routine produce successful simulation of routine produce successful simulation of routine produce successful simulation of routine commonsense cognitive competences is commonsense cognitive competences is notorious, not to say scandalous. We still notorious, not to say scandalous. We still don't have the fabled machine that can don't have the fabled machine that can make breakfast without burning down the make breakfast without burning down the house; or the one that can translate house; or the one that can translate everyday English into everyday Italian, or everyday English into everyday Italian, or the one that can summarize texts..” the one that can summarize texts..” (Jerry Fodor, The Mind doesn’t Work (Jerry Fodor, The Mind doesn’t Work that Way, 2000, p.37).that Way, 2000, p.37).

“ The whole of science is nothingmore than a refinement of everydaythinking”.

- Albert Einstein- Albert Einstein

ReferencesReferences

�� 71. S.M.Aqil Burney; Adnan Manzoor; Fahad Burney. Similarities and Differences in 71. S.M.Aqil Burney; Adnan Manzoor; Fahad Burney. Similarities and Differences in Statistics and Data Mining. Accepted for Publication at International Workshop on Statistics and Data Mining. Accepted for Publication at International Workshop on Frontiers of Information Technology Islamabad Pakistan organized by COMSATS Frontiers of Information Technology Islamabad Pakistan organized by COMSATS Institute of Information [email protected] of Information [email protected]

�� 72. S.M.Aqil Burney; Tahseen A. Jilani Supervised Artificial Neural Network for 72. S.M.Aqil Burney; Tahseen A. Jilani Supervised Artificial Neural Network for Karachi Stock Exchange Share Rates Forecasting” Accepted at Artificial Intelligence Karachi Stock Exchange Share Rates Forecasting” Accepted at Artificial Intelligence and Neural Network Conference .Department of Computer Engineering Canakkle and Neural Network Conference .Department of Computer Engineering Canakkle University ,TurkeyJuly2University ,TurkeyJuly2--4,20034,2003

�� TAINNTAINN--2003.2003.�� TAINNTAINN--2003.2003.

�� 73. S.M.Aqil Burney(2003) Introduction to Stochastic Processes and their 73. S.M.Aqil Burney(2003) Introduction to Stochastic Processes and their Applications in Insurance and Computer Science. Dr.M.N.M. Talpur Memorial Lecture. Applications in Insurance and Computer Science. Dr.M.N.M. Talpur Memorial Lecture. Published by Institute of Mathematics and Computer Science University of Sindh, Published by Institute of Mathematics and Computer Science University of Sindh, Jamshoro.Jamshoro.

�� 20042004�� 74. S.M.Aqil Burney; Asadullah Malik. QOS Management for Mobile Internet Services 74. S.M.Aqil Burney; Asadullah Malik. QOS Management for Mobile Internet Services

by Optimization of System Efficiency , Capacity , Production Ratio and Handoff Rate.by Optimization of System Efficiency , Capacity , Production Ratio and Handoff Rate.

�� 75.S.M.Aqil Burney, Tahseen A. Jilani & Cemal Ardil(2004)Levenberg75.S.M.Aqil Burney, Tahseen A. Jilani & Cemal Ardil(2004)Levenberg--Marquardt Marquardt Algorithm for Karachi Stock Exchange Share Rate Forecasting .Journal of Algorithm for Karachi Stock Exchange Share Rate Forecasting .Journal of Computational Intelligence. Canakkale University, Turkey.Computational Intelligence. Canakkale University, Turkey.

�� 20052005�� 82. Aqil Burney, S.M. et al, (2005), A Brief History of Logic: Application of Logic in 82. Aqil Burney, S.M. et al, (2005), A Brief History of Logic: Application of Logic in

Computer Science and Information Technology, in proc. of the National Conference on Computer Science and Information Technology, in proc. of the National Conference on Information Technology and Applications (NCITA) April 21Information Technology and Applications (NCITA) April 21--22,2005, Balochistan 22,2005, Balochistan University of Information Technology and Management Sciences, Quetta, Pakistan.University of Information Technology and Management Sciences, Quetta, Pakistan.

�� 83. Aqil Burney, S.M. et al,(2005), Approximate Bounded Knowledge Extraction using 83. Aqil Burney, S.M. et al,(2005), Approximate Bounded Knowledge Extraction using TypeType--I Fuzzy Logic, in proc. of the 5th World Enformatika Conference, WEC’05, August I Fuzzy Logic, in proc. of the 5th World Enformatika Conference, WEC’05, August 2626--28, 2005 in Prague, Czech Republic. ISBN: 97528, 2005 in Prague, Czech Republic. ISBN: 975--9845898458--66--5. ISSN (e): 13055. ISSN (e): 1305--5313.5313.

�� 84. Aqil Burney, S.M. et al,(2005), Approximate Bounded Knowledge Extraction using 84. Aqil Burney, S.M. et al,(2005), Approximate Bounded Knowledge Extraction using TypeType--I Fuzzy Logic, in proc. of the 5th World Enformatika Conference, WEC’05, August I Fuzzy Logic, in proc. of the 5th World Enformatika Conference, WEC’05, August 2626--28, 2005 in Prague, Czech Republic. ISBN: 97528, 2005 in Prague, Czech Republic. ISBN: 975--9845898458--66--5. ISSN (e): 13055. ISSN (e): 1305--5313.5313.

�� 2006200620062006�� 85. Aqil Burney, S.M. et al,(2006), Approximate Knowledge Extraction Using Multi85. Aqil Burney, S.M. et al,(2006), Approximate Knowledge Extraction Using Multi--

resolution Analysis for Typeresolution Analysis for Type--I Fuzzy Neural Networks, published in ICEASI Fuzzy Neural Networks, published in ICEAS--2006, 2006, Islamabad, Pakistan. PP. 464Islamabad, Pakistan. PP. 464--469. accessible through IEEE469. accessible through IEEE--Xplore, USA.Xplore, USA.

�� 86. Aqil Burney, S.M. et al, (2006), Wavelet Based SARIMA Models for Forecasting 86. Aqil Burney, S.M. et al, (2006), Wavelet Based SARIMA Models for Forecasting Neural Gas Demand. Business Review Vol.I, No.1, ISSN 1990Neural Gas Demand. Business Review Vol.I, No.1, ISSN 1990--6587, IBA, Karachi6587, IBA, Karachi--Pakistan. PP 134Pakistan. PP 134--139139

�� 20072007�� 87. Aqil Burney S. M., Jilani T. A. and Ardil C. (2007a), “Multivariate High Order Fuzzy 87. Aqil Burney S. M., Jilani T. A. and Ardil C. (2007a), “Multivariate High Order Fuzzy

Time Series Forecasting for Car Road Accidents”, International Journal of Computational Time Series Forecasting for Car Road Accidents”, International Journal of Computational Intelligence, 4(1), pp 15Intelligence, 4(1), pp 15--20, ISSN 130420, ISSN 1304--2386.2386.

�� 88. Aqil Burney S. M. and Jilani T. A. (2007), M88. Aqil Burney S. M. and Jilani T. A. (2007), M--Factor High Order Fuzzy Time Series Factor High Order Fuzzy Time Series Forecasting for Road Accident Data, Accepted in IEEEForecasting for Road Accident Data, Accepted in IEEE--IFSA, 18IFSA, 18--21 June, Cancun21 June, Cancun--Mexico. Published in Analysis and Design of Intelligent Systems Using Fuzzy Logic and Mexico. Published in Analysis and Design of Intelligent Systems Using Fuzzy Logic and Soft Computing, Advances in Soft Computing, 41, SpringerSoft Computing, Advances in Soft Computing, 41, Springer--Verlag.Verlag.

�� 89. Aqil Burney S. M. and Jilani T. A. and (2007), New Method of Learning and 89. Aqil Burney S. M. and Jilani T. A. and (2007), New Method of Learning and Knowledge Management in TypeKnowledge Management in Type--I Fuzzy Neural Networks, Accepted in IEEEI Fuzzy Neural Networks, Accepted in IEEE--IFSA, 18IFSA, 18--21 21 June, World Congress, CancunJune, World Congress, Cancun-- Mexico. Published in Theoretical Advances and Mexico. Published in Theoretical Advances and Applications of Fuzzy Logic and Soft Computing; Advances in Soft Computing, 42, Applications of Fuzzy Logic and Soft Computing; Advances in Soft Computing, 42, SpringerSpringer--Verlag.Verlag.

�� 90. Aqil Burney S. M. and Jilani T. A. (2007), Multivariate Stochastic Fuzzy Forecasting 90. Aqil Burney S. M. and Jilani T. A. (2007), Multivariate Stochastic Fuzzy Forecasting Models, In Expert Systems with Applications, ESWA, to apppear in 37(2) in late 2008 or Models, In Expert Systems with Applications, ESWA, to apppear in 37(2) in late 2008 or 2009.2009.

�� 91. Aqil Burney S. M., Jilani T. A. and Ardil C. (2007), Fuzzy Metric approach for Fuzzy 91. Aqil Burney S. M., Jilani T. A. and Ardil C. (2007), Fuzzy Metric approach for Fuzzy time Series Forecasting Based on Frequency Density Based Partitioning, Proceedings of time Series Forecasting Based on Frequency Density Based Partitioning, Proceedings of World Academy of Science, Engineering and Technology vol. 23, August, ISSN 1307World Academy of Science, Engineering and Technology vol. 23, August, ISSN 1307--68846884

�� 92. S. M. Aqil Burney , Akhtar Raza (2007), Monte Carlo Simulation and Prediction of 92. S. M. Aqil Burney , Akhtar Raza (2007), Monte Carlo Simulation and Prediction of internet load using conditional mean and conditional variance model. Proceedings of The internet load using conditional mean and conditional variance model. Proceedings of The 9th Islamic Countries Conference on Statistical Sciences (ICCS9th Islamic Countries Conference on Statistical Sciences (ICCS--IX) 12IX) 12--14 December 14 December 20072007

�� 20082008�� 93. Aqil Burney S. M. and Jilani T. A. (2008), A Refined Fuzzy 93. Aqil Burney S. M. and Jilani T. A. (2008), A Refined Fuzzy

Time Series Model for Stock Market Forecasting. PHYSICA A: Time Series Model for Stock Market Forecasting. PHYSICA A: Statistical Mechanics with Applications(Econophysics and Models) Statistical Mechanics with Applications(Econophysics and Models) Volume 387, Issue 12, 1 May 2008, Pages 2857Volume 387, Issue 12, 1 May 2008, Pages 2857--2862, Elsevier 2862, Elsevier Publishers.Publishers.

�� 94. Aqil Burney S. M. and Jilani T. A. (2008), Multivariate 94. Aqil Burney S. M. and Jilani T. A. (2008), Multivariate �� 94. Aqil Burney S. M. and Jilani T. A. (2008), Multivariate 94. Aqil Burney S. M. and Jilani T. A. (2008), Multivariate Stochastic Fuzzy Forecasting Models. Expert Systems with Stochastic Fuzzy Forecasting Models. Expert Systems with Applications: An International Journal, Volume 35, Issue 3 Applications: An International Journal, Volume 35, Issue 3 (October 2008) Pages 691(October 2008) Pages 691--700. ISSN:0957700. ISSN:0957--41744174

�� 95. Tahseen Ahmed Jilani, Syed Muhammad Aqil Burney, C. Ardil 95. Tahseen Ahmed Jilani, Syed Muhammad Aqil Burney, C. Ardil (2008), Fuzzy Metric Approach for Fuzzy Time Series Forecasting (2008), Fuzzy Metric Approach for Fuzzy Time Series Forecasting based on Frequency Density Based Partitioning. International based on Frequency Density Based Partitioning. International Journal of Computation (Internet), World Academy of Science, Journal of Computation (Internet), World Academy of Science, Engineering and Technology (Volume 4 Number 2 Spring 2008) Engineering and Technology (Volume 4 Number 2 Spring 2008) Pages 112Pages 112--117 www.waset.org 117 www.waset.org