Intelligent Decision Making: An AI-Based Approach

410 Pages·2008·12.23 MB·English

by  Gloria Phillips-Wren

Gloria Phillips-Wren, Nikhil Ichalkaranje and Lakhmi C. Jain (Eds.)
Intelligent Decision Making: An AI-Based Approach
StudiesinComputationalIntelligence,Volume97
Editor-in-chief
Prof.JanuszKacprzyk
SystemsResearchInstitute
PolishAcademyofSciences
ul.Newelska6
01-447Warsaw
Poland
E-mail:[email protected]
Furthervolumesofthisseriescanbefoundonour Vol.86.ZbigniewLesandMogdalenaLes
homepage:springer.com ShapeUnderstandingSystems,2008
ISBN978-3-540-75768-9
Vol.74.RobertSchaefer Vol.87.YuriAvramenkoandAndrzejKraslawski
FoundationofGlobalGeneticOptimization,2007 CaseBasedDesign,2008
ISBN978-3-540-73191-7 ISBN978-3-540-75705-4
Vol.75.CrinaGrosan,AjithAbrahamandHisaoIshibuchi Vol.88.TinaYu,DavidDavis,CemBaydarandRajkumar
(Eds.) Roy(Eds.)
HybridEvolutionaryAlgorithms,2007 EvolutionaryComputationinPractice,2008
ISBN978-3-540-73296-9 ISBN978-3-540-75770-2
Vol.76.SubhasChandraMukhopadhyayandGourabSen Vol.89.ItoTakayuki,HattoriHiromitsu,ZhangMinjie
Gupta(Eds.) andMatsuoTokuro(Eds.)
AutonomousRobotsandAgents,2007 Rational,Robust,Secure,2008
ISBN978-3-540-73423-9 ISBN978-3-540-76281-2
Vol.77.BarbaraHammerandPascalHitzler(Eds.) Vol.90.SimoneMarinaiandHiromichiFujisawa(Eds.)
PerspectivesofNeural-SymbolicIntegration,2007 MachineLearninginDocumentAnalysis
ISBN978-3-540-73953-1 andRecognition,2008
Vol.78.CostinBadicaandMarcinPaprzycki(Eds.) ISBN978-3-540-76279-9
IntelligentandDistributedComputing,2008
Vol.91.HorstBunke,KandelAbrahamandLastMark(Eds.)
ISBN978-3-540-74929-5
AppliedPatternRecognition,2008
Vol.79.XingCaiandT.-C.JimYeh(Eds.) ISBN978-3-540-76830-2
QuantitativeInformationFusionforHydrological
Vol.92.AngYang,YinShanandLamThuBui(Eds.)
Sciences,2008
SuccessinEvolutionaryComputation,2008
ISBN978-3-540-75383-4
ISBN978-3-540-76285-0
Vol.80.JoachimDiederich
Vol.93.ManolisWallace,MariosAngelidesandPhivos
RuleExtractionfromSupportVectorMachines,2008
Mylonas(Eds.)
ISBN978-3-540-75389-6
AdvancesinSemanticMediaAdaptationand
Vol.81.K.Sridharan Personalization,2008
RoboticExplorationandLandmarkDetermination,2008 ISBN978-3-540-76359-8
ISBN978-3-540-75393-3
Vol.94.ArpadKelemen,AjithAbrahamandYuehuiChen
Vol.82.AjithAbraham,CrinaGrosanandWitold (Eds.)
Pedrycz(Eds.) ComputationalIntelligenceinBioinformatics,2008
EngineeringEvolutionaryIntelligentSystems,2008 ISBN978-3-540-76802-9
ISBN978-3-540-75395-7 Vol.95.Radu Dogaru
Vol.83.BhanuPrasadandS.R.M.Prasanna(Eds.) Systematic Design for Emergence in Cellular Nonlinear
Speech,Audio,ImageandBiomedicalSignalProcessing Networks, 2008
usingNeuralNetworks,2008 ISBN978-3-540-76800-5
ISBN978-3-540-75397-1 Vol.96.Aboul Ella Hassanien, Ajith Abraham and Janusz
Vol.84.MarekR.OgielaandRyszardTadeusiewicz Kacprzyky(Eds.)
ModernComputationalIntelligenceMethods Computational Intelligence in Multimedia Processing:
fortheInterpretationofMedicalImages,2008 Recent Advances,2008
ISBN978-3-540-75399-5 ISBN978-3-540-76826-5
Vol.85.ArpadKelemen,AjithAbrahamandYulanLiang Vol.97.Gloria Phillips-Wren, Nikhil Ichalkaranje and
(Eds.) Lakhmi C. Jain (Eds.)
ComputationalIntelligenceinMedicalInformatics,2008 IntelligentDecision Making: An AI-Based Approach,2008
ISBN978-3-540-75766-5 ISBN978-3-540-76829-9
Gloria Phillips-Wren
Nikhil Ichalkaranje
Lakhmi C. Jain
(Eds.)
Intelligent Decision Making:
An AI-Based Approach
With107Figuresand44Tables
123
Prof. Dr. Gloria Phillips-Wren Dr. Nikhil Ichalkaranje
Information Systems and Operations Management School of Electrical
The Sellinger School of Business and Management and Information Engineering
Loyola College in Maryland University of South Australia
4501 N. Charles Street Adelaide
Baltimore, MD 21210 South Australia SA 5095
USA Australia
Prof. Dr. Lakhmi C. Jain
School of Electrical and Information Engineering
University of South Australia
Adelaide
South Australia SA 5095
Australia
ISBN978-3-540-76828-9 e-ISBN978-3-540-76829-6
StudiesinComputationalIntelligenceISSN1860-949X
LibraryofCongressControlNumber:2007939887
(cid:1)c 2008Springer-VerlagBerlinHeidelberg
Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerial
isconcerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation,broad-
casting,reproductiononmicrofilmorinanyotherway,andstorageindatabanks.Duplicationof
thispublicationorpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLaw
ofSeptember9,1965,initscurrentversion,andpermissionforusemustalwaysbeobtainedfrom
Springer-Verlag.ViolationsareliabletoprosecutionundertheGermanCopyrightLaw.
Theuseofgeneraldescriptivenames,registerednames,trademarks,etc.inthispublicationdoesnot
imply, even in the absence of a specific statement, that such names are exempt from the relevant
protectivelawsandregulationsandthereforefreefor general use.
CoverDesign:Deblik,Berlin,Germany
Printedonacid-freepaper
9 8 7 6 5 4 3 2 1
springer.com
Dedicated to our mothers who were our first and best teachers.
Preface
Thefusionofartificialintelligence(AI)withdecisionsupportsystems(DSSs)
is opening exciting new areas of research and application. The resulting sys-
tems are smarter, more efficient, adaptable, and better able to aid human
decision making. While AI aims to mimic human behaviour in limited ways,
DSSs attempt to help humans make the best choice among a set of possible
choices given explicit or implied criteria. Long a topic of science fiction, AI
today is demonstrating that it can be integrated effectively into real systems
and that it offers the only way possible to capture aspects of human intelli-
gence such as learning.The combinationof AI and DSSs provides formidable
new computational assistants to humans that extend their capabilities in
routine and complex stressful environments. Due to the increasing matu-
rity of this interdisciplinary field as evidenced by the recent growth in the
number of research publications and contributors entering the field, a book
that explores the current state and future outlook of intelligent DSSs seems
appropriate.
Thebookisorganizedaroundthreethemes.Thefirsttwochaptersprovide
asolidfoundationbyexploringstudiesandtheoriesofhumandecisionmaking.
They trace some one hundred years of research including recent work by the
well-known authors and provide a vision of the use of computerized decision
aids.ThesecondsectiondealswithparadigmsandmethodsassociatedwithAI
in DSS. The final section provides sample applications among the many that
are appearing today and gives our perspective on future research directions
needed to advance the field.
Thisbookwouldnothavebeenpossiblewithouttheeffortsofmanypeople.
We thank the contributors for their inspiring research and the reviewers for
their efforts to create a high-quality book. The publisher’s support, patience
andassistancearegratefullyacknowledged.Inparticular,SrilathaAchuthan’s
unwaveringeffortsasprojectmanagerprovidedhelpwhenweneededitmost.
VIII Preface
We thankthe researchcommunityforthe advancesthathavemadethis book
possible and our families for their continued support.
USA Gloria Phillips-Wren
Australia Nikhil Ichalkaranje
Australia Lakhmi C. Jain
Foreword
Intelligentdecisionsystems(IDS)arearelativelynewparadigminthedecision
supportsystems(DSS) area.Consistentwiththe modernviewonworkactiv-
ityasmostly‘knowledgework’(Davenport,2005)andrecognisingthe critical
role of knowledge for effective decision-making, intelligent decision support
aims to provide the decision maker with quality assistance in gaining better
knowledge and understanding of the decisionsituation. IDS are the means to
achieve such assistance.
This needfor knowledgemanagementandprocessingwithin decisionsup-
port systems has resulted in a special class of systems that incorporates
qualitative knowledge and reasoning, extending the functionality beyond
those traditionally covered by DSS applications. These systems, variously
termed Intelligent Decision Support Systems, Intelligent Decision Systems,
Knowledge-BasedDecisionSupportSystems,ActiveDSSandJointCognitive
Systems, include qualitative knowledge to extend the typically quantitative
data ofearlierapproachesto decisionsupport(BursteinandHolsapple,2008;
Gupta et al. 2006).
The label intelligent in IDS is derived from the attempts made in artifi-
cial intelligence (AI) to develop systems that computationally emulate some
human cognitive capabilities such as reasoning, learning and memory. The
need to incorporate domain knowledge and intelligent capabilities in deci-
sion support systems has been identified in various forms and models by
many researchers, starting from Simon (1977), followed by Sprague (1993),
and exemplified by Turban, Aronson and Liang (2005) and Holsapple and
Whinston (1996) in their comprehensive analyses of tools and techniques for
incorporatingintelligenceintoDSS.ArnottandPervan(2005),intheirreview
of the DSS field, traced and described Intelligent Decision Support as a sep-
arate branch, which originated from research in AI and Expert Systems to
complement the needs of modern PersonalisedDecision Support.
The main role of IDS in an organisation is as an enabler for knowledge
processingwithcommunicationcapabilitiestosupportknowledgesharingand
exchangeandtofacilitateorganisationallearning(CarlssonandKalling,2006;
X Foreword
Bursteinand Linger,2003).IDS aimto assistthe decisionmaker in overcom-
ingcognitivelimitationstoachievingthebestdecisionoutcomes.Atthesame
time the system could identify some useful knowledge for future improve-
ments in the decision-making process, thus facilitating continuous learning
processes by an organisation.ConventionalDSS was not intended to support
such functionality, hence giving rise to IDS in a knowledge management con-
text. Despite the significant potential of IDS and remarkable advances in AI
technologies, the promise of IDS has not yet been realized.
IDSarenotwidespreadassuch.Onereasonisthatcomprehensiveresearch
is still required on AI technologies to be used in IDS. Some technologies
such as intelligent agents have advanced to the point that they are imple-
mented in numerous practical applications, while other AI concepts such as
neural networks are not yet as mature. In most cases, specialized IDS appli-
cations are reported in the literature, although generalized applications have
notbeendeveloped.Researchisneededonarchitecturesandframeworksthat
could support production-level IDS both at the AI and at systems levels.
Although IDS do not in general exist as stand-alone systems, any large-scale
management information system would include some intelligent components.
Modern approaches to assisting organizations such as customer relationship
management (CRM), knowledge management systems (KMS), and business
intelligence (BI) systems are heavily influenced by intelligent techniques and
include a wide range of intelligent systems functionality. Many such systems
requireaccesstoexpertorproblem-domainknowledge.Availabilityofsophis-
ticated generic technological infrastructure makes it easier to specialise such
systems to suit specific application domains.
A number of books have been published in the area of IDS and related
areas of Intelligent Decision Support Systems, and one needs to ask what
another book can add to the community. Publication patterns over the last
10years(showninFig.1basedondatafromGoogleScholar)appeartoshow
continuedinterestinIDS.Thisisamuchneededbooktoupdatetheinterested
reader in an exciting research field with many opportunities for advances in
both theoretical and applied areas.
The current volume is an effort to bridge the range of exploration in
this field from fundamental understanding of human decision making at an
abstract conceptual level, to methods of computational intelligence, and to
applicationsofintelligentdecisionsupporttechniquesinspecificcontexts.The
bookpresentsfascinatingbackgroundinformationonhumandecisionmaking
and makes a contribution to the IDS area by presenting the current state of
knowledgeand identifying key researchgaps.I would like to congratulatethe
editors of this book and look forward to it being remembered as a pivotal
beginning for collective focus and mutual inspiration.
Victoria, Australia Frada V. Burstein
Foreword XI
25,000
20,000
15,000 intelligent decision support
intelligent decision support
systems
10,000
intelligent decision systems
5,000
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Fig.1.ComparativedataonpublicationsinIntelligentDecisionSupport,Intelligent
DecisionSupportSystemsandIntelligentDecisionSystems(basedonthedatafrom
Google Scholar)
References
Arnott,D.andPervan,G.(2005). ACritical Analysisof Decision SupportSystems
Research. Journal of Information Technology, 20, 67–87.
Burstein, F. and Holsapple, C.W. (eds.) (2008). Handbook on Decision Support
Systems. Berlin/Heidelberg: Springer.
Burstein, F. and Linger, H. (2003). Supporting Post-Fordist Work Practices: A
KnowledgeManagement FrameworkforSupportingKnowledgeWork.Informa-
tion Technology and People, 16, 3, 289–305.
Carlsson, S.A. and Kalling, T. (2006). Decision Support through Knowledge Man-
agement: What Works and What Breaks. In Adam, F., Br´ezillon, P., Carlsson,
S.andHumphreys,P.(eds.),Creativity and Innovation inDecisionMaking and
Decision Support, London, UK:Decision Support Press, 693–710.
Davenport,T.H.(2005).ThinkingforaLiving:HowtoGetBetterPerformanceand
Results from Knowledge Workers. Boston, MA: Harvard Business School Press.
GuptaJ.,ForgionneG.,andMoraM.(eds.)(2006).IntelligentDecision-MakingSup-
port Systems (i-DMSS): Foundations, Applications and Challenges.Engineering
Decision Series of Springer.
Simon, H. (1977). The New Science of Management Decisions. New Jersey, NJ:
Prentice-Hall.
Sprague, R.H. (1993). A Framework for the Development of Decision Support Sys-
tems. In Sprague, R.H. and Watson, H.J. (eds.), Decision Support Systems
PuttingTheoryintoPractice,NewJersey,NJ:Prentice-HallInternational,3–26.
Turban, E., Aronson, J. and Liang, T. (2005). Decision Support Systems and
Intelligence Systems. 7th Ed., New Jersey: Pearson Prentice-Hall.
Holsapple C. and Whinston, A. (1996). Decision Support Systems: A Knowledge
Based Approach. St.Paul, MN:West Publishing.