Computational Intelligence for Agent-based Systems

274 Pages·2007·3.12 MB·English

by  Raymond S.T. Lee

RaymondS.T.LeeandVincenzoLoia(Eds.)
ComputationalIntelligenceforAgent-basedSystems
StudiesinComputationalIntelligence,Volume72
Editor-in-chief
Prof.JanuszKacprzyk
SystemsResearchInstitute
PolishAcademyofSciences
ul.Newelska6
01-447Warsaw
Poland
E-mail:[email protected]
Furthervolumesofthisseries Vol.63.WlodzislawDuchandJacekMan´dziuk(Eds.)
canbefoundonourhomepage: ChallengesforComputationalIntelligence,2007
springer.com ISBN978-3-540-71983-0
Vol.64.LorenzoMagnaniandPingLi(Eds.)
Vol.53.HuajinTang,KayChenTan,ZhangYi
Model-BasedReasoninginScience,Technology,and
NeuralNetworks:ComputationalModels
Medicine,2007
andApplications,2007
ISBN978-3-540-71985-4
ISBN978-3-540-69225-6
Vol.54.FernandoG.Lobo,Cla´udioF.Lima Vol.65.S.Vaidya,L.C.JainandH.Yoshida(Eds.)
andZbigniewMichalewicz(Eds.) AdvancedComputationalIntelligenceParadigmsin
ParameterSettinginEvolutionaryAlgorithms,2007 Healthcare-2,2007
ISBN978-3-540-69431-1 ISBN978-3-540-72374-5
Vol.55.XianyiZeng,YiLi,DaRuanandLudovicKoehl Vol.66.LakhmiC.Jain,VasilePaladeandDipti
(Eds.) Srinivasan(Eds.)
ComputationalTextile,2007 AdvancesinEvolutionaryComputingforSystem
ISBN978-3-540-70656-4 Design,2007
ISBN978-3-540-72376-9
Vol.56.AkiraNamatame,SatoshiKuriharaand
HideyukiNakashima(Eds.) Vol.67.VassilisG.KaburlasosandGerhardX.Ritter
EmergentIntelligenceofNetworkedAgents,2007 (Eds.)
ISBN978-3-540-71073-8 ComputationalIntelligenceBasedonLattice
Theory,2007
Vol.57.NadiaNedjah,AjithAbrahamandLuizade
ISBN978-3-540-72686-9
MacedoMourella(Eds.)
ComputationalIntelligenceinInformationAssurance
andSecurity,2007 Vol.68.CiprianoGalindo,Juan-Antonio
ISBN978-3-540-71077-6 Ferna´ndez-MadrigalandJavierGonzalez
AMulti-HierarchicalSymbolicModel
Vol.58.Jeng-ShyangPan,Hsiang-ChehHuang,Lakhmi oftheEnvironmentforImprovingMobileRobot
C.JainandWai-ChiFang(Eds.) Operation,2007
IntelligentMultimediaDataHiding,2007 ISBN978-3-540-72688-3
ISBN978-3-540-71168-1
Vol.69.FalkoDresslerandIacopoCarreras(Eds.)
Vol.59.AndrzejP.WierzbickiandYoshiteru AdvancesinBiologicallyInspiredInformationSystems:
Nakamori(Eds.) Models,Methods,andTools,2007
CreativeEnvironments,2007 ISBN978-3-540-72692-0
ISBN978-3-540-71466-8
Vol.70.JavaanSinghChahl,LakhmiC.Jain,Akiko
Vol.60.VladimirG.IvancevicandTijanaT.Ivacevic
MizutaniandMikaSato-Ilic(Eds.)
ComputationalMind:AComplexDynamics
InnovationsinIntelligentMachines-1,2007
Perspective,2007
ISBN978-3-540-72695-1
ISBN978-3-540-71465-1
Vol.61.JacquesTeller,JohnR.LeeandCatherine Vol.71.NorioBaba,LakhmiC.JainandHisashiHanda
Roussey(Eds.) (Eds.)
OntologiesforUrbanDevelopment,2007 AdvancedIntelligentParadigmsinComputer
ISBN978-3-540-71975-5 Games,2007
ISBN978-3-540-72704-0
Vol.62.LakhmiC.Jain,RaymondA.Tedman
andDebraK.Tedman(Eds.) Vol.72.RaymondS.T.LeeandVincenzoLoia(Eds.)
EvolutionofTeachingandLearningParadigms ComputationalIntelligenceforAgent-based
inIntelligentEnvironment,2007 Systems,2007
ISBN978-3-540-71973-1 ISBN978-3-540-73175-7
Raymond S.T. Lee
Vincenzo Loia
(Eds.)
Computational Intelligence
for Agent-based Systems
With 82 Figures and 11 Tables
ProfessorDr.RaymondS.T.Lee ProfessorDr.VincenzoLoia
DepartmentofComputing DipartimentodiMatematicaeInformatica
TheHongKongPolytechnicUniversity UniversitadiSalerno
Kowloon,HungHom viaPontedonMelillo
HongKong 84084Fisciano(Salerno)
E-mail:[email protected] Italy
E-mail:[email protected]
LibraryofCongressControlNumber:2007929547
ISSNprintedition:1860-949X
ISSNelectronicedition:1860-9503
ISBN978-3-540-73175-7SpringerBerlinHeidelbergNewYork
Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerial
isconcerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation,broad-
casting,reproductiononmicrofilmorinanyotherway,andstorageindatabanks.Duplicationof
thispublicationorpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLaw
ofSeptember9,1965,initscurrentversion,andpermissionforusemustalwaysbeobtainedfrom
Springer-Verlag.ViolationsareliabletoprosecutionundertheGermanCopyrightLaw.
SpringerisapartofSpringerScience+BusinessMedia
springer.com
(cid:176)c Springer-VerlagBerlinHeidelberg2007
Theuseofgeneraldescriptivenames,registerednames,trademarks,etc.inthispublicationdoesnot
imply, even in the absence of a specific statement, that such names are exempt from the relevant
protectivelawsandregulationsandthereforefreeforgeneraluse.
Coverdesign:deblik,Berlin
TypesettingbytheSPiusingaSpringerLATEXmacropackage
Printedonacid-freepaper SPIN:11790952 89/SPi 543210
Foreword
The scope of this volume is to give to the reader a wide scenario of recent
works characterized by a synergistic combination of Soft Computing area
with recent trends of Distributed Artificial Intelligence and Ambient Intel-
ligence. We edited this book for two basic motivations: the emergence of
computational intelligence as a mature and integrated science, and the power
of the agent paradigm in realizing complex and distributed environments.
Computational Intelligence has nowadays a mature formal theory and robust
implementation experiences. Intelligent agents have shown their inner ability
in acquiring data and knowledge from and performing appropriate deci-
sions through a functional relation that maps percepts to actions. Computa-
tional Intelligence inside Agents are extremely helpful to support the designer
of agent-based systems in treating with unknown or ill-defined environ-
ments so to better handle knowledge representation and reasoning under
these assumptions. The formulation of cooperative strategies, the linguistic
aspects of the communication, the local decision versus the global knowl-
edge, the actions and plans as distributed viewpoints could be characterized
by an incomplete knowledge and by a fragmentation of logical interac-
tions, faced often by a non-monotonic reasoning.
While building “smart” environments many challenges arises. The system
needs to interact with its users in order to obtain feedback in a non-intrusive
way. Explicit/implicit feedback is needed to take decisions in near-real time.
To reach this objective, agents have many advantages. The agents communi-
cate with one another by exchanging asynchronous or synchronous, goal-
driven messages. Decision making and learning strategies are facilitated in
that each agent has knowledge just of a small part of the environment and
takes decisions only about this sub-view. Decentralization and concurrent
execution allow to better respect the real time commitment. Respecting
this trend, widely recognized in the design of Ambient Intelligence sys-
tems, this book proposes contributions where Computational Intelligence
plays an important role. The emphasis of Ambient Intelligence is on greater
user-friendliness, more efficiency in services’ support, user empowerment
and support for human interactions. People are surrounded by smart pro-
active devices that are embedded in all kinds of objects: an intelligent
vi Foreword
environment is capable of recognizing and responding to the presence of
different individuals, working in a seamless, unobtrusive and often invisi-
ble way.
This book explores these emerging areas inviting well-known authors
whose expertise is widely recognized.
In Chapter 1, Acampora and Loia combine together Computational Intelli-
gence methodologies and Semantic Web technologies to develop a general
framework where ubiquitous devices are able to find the suitable set of‘intel-
ligent’ services in a transparent way.
In Chapter 2, Costa-Montenegro et al. evaluate the feasibility of intelli-
gent and distributed control of shared resources in user-managed networks.
Their study presents as main conclusion that a user network can obtain bene-
fits from node cooperation if traffic demands are variable along the day.
Besides, the selection of agent-based management with individual credit
limits provides the best results for every user, and also for the whole net-
work, as cooperation emerges as the preferred strategy.
Chapter 3, by Duman et al., describes the project iDorm, an experimental
test bed for ubiquitous computing environments research at Essex University.
The iDorm is equipped with a number of passive and intelligent embedded
agents with different capabilities and restrictionsIn the chapter the authors
define a passive agent as an embedded computer that delivers purely sensory
information, whereas the intelligent embedded agents are embedded-
computers enhanced with intelligent reasoning and learning.
Lam and Lee, in Chapter 4, present iJADE FreeWalker, a tourist context-
aware guiding system, which is constructed by using Semantic Web techno-
logies. iJADE FreeWalker integrates GPS, ontology and agent technologies
to support location awareness for providing the precise navigation and clas-
sify the tourist information for the users. The system was tested with success
in a real application that underlined how 83% of the users felt that the sys-
tem can help tourists find tourist information in Hong Kong.
Chapter 5, by, Lim and Lee, illustrates iJADE InfoSeeker, an intelligent
context-aware agents system that is designed to help users find, retrieve,
and analyze news article from the Internet and then present the content in a
semantic web. The authors, Lim and Lee, discuss the advantages of using
multiple intelligent agents to mine news articles on the web, underlining
the benefits of using ontologies to analyze the semantics of Chinese text,
and also the advantages of using a semantic web to organize information
semantically.
Foreword vii
Chapter 6, written by Martin et al., reports the scientific results and experi-
ences coming from a complex and a concrete project on ambient intelli-
gence employed for third generation telecare systems. The experiment has
been conducted in homes of elderly clients over a long term and is able to
detect a person’s movements and use of furniture and household items.
Murthy and Krishnamurthy discuss in Chapter 7 how a multi-agent sys-
tem can model the evolution of a Nature-Inspired Smart System exhibiting
four salient properties: (i) Collective, coordinated and efficient (ii) Self-
organization and emergence (iii) Power law scaling or scale invariance under
emergence (iv) Adaptive, fault tolerant and resilient against damage.
Seif El-Nasr and Vasilakos propose, in Chapter 8, intelligent systems
that use non-linear optimization as well as symbolic rule-based systems to
adjust sound/music, on-stage lighting, and projected visual imagery in terms
of movement and colour to dynamically reflecting the dancer’s arousal state
measured through physiological sensors worn by the dancers.
We deeply thank all the invited authors. A particular acknowledgement
goes to Prof. J. Kacprzyk, the Editor-in-Chief of this Series, which has given
to us the opportunity to act as Guest Editors of this book.

The Guest Editors,

Raymond S.T. Lee, The Hong Kong Polytechnic University
Vincenzo Loia, University of Salerno
Contents
Foreword....................................................................................................v
List of Contributors.................................................................................xv
A Proposal of an Open Ubiquitous Fuzzy Computing System
for Ambient Intelligence............................................................................1
Abstract...................................................................................................1
1.1 Introduction....................................................................................1
1.2 AmI Fuzzy Ubiquitous System Architecture.................................3
1.3 AmI Context Representation..........................................................5
1.4 Transparent Fuzzy Control for AmI Context Representation........6
1.4.1 FML and Transparent Fuzzy Control....................................8
1.5 Run Time Environment................................................................11
1.5.1 Run Time Servers Technologies..........................................12
1.5.2 AmI Clients..........................................................................14
1.5.3 Client/Server Communication Protocol...............................14
1.6 Fuzzy Services Retrieval..............................................................15
1.6.1 AmI Client Sensor/Actuator Network Knowledge..............18
1.6.2 Storing Algorithm................................................................19
1.6.3 Retrieval Algorithm.............................................................23
1.7 Final Considerations.....................................................................25
References............................................................................................26
Agent-Controlled Sharing of Distributed Resources in User Networks 29
Abstract................................................................................................29
2.1 Introduction..................................................................................30
2.2 User-Managed Networks .............................................................32
2.3 Peer-to-Peer Models.....................................................................33
2.4 Game Theory................................................................................35
2.5 Multiagent Systems......................................................................36
2.6 Strategies, Node Topology and Behavior ....................................37
2.7 A Simple Cellular Automata Approach.......................................40
2.8 A Basic Multi-Agent Approach...................................................44
2.8.1 Strategy Types ....................................................................44
x Contents
2.8.2 Network Traffic..................................................................45
2.8.3 Simulation Scenarios..........................................................45
2.8.4 Learning CreditLimit..........................................................46
2.8.5 Simulation Results..............................................................47
2.9 A More Realistic Multi-Agent Approach.....................................49
2.9.1 Considerations on User Network Topology and Traffic.....49
2.9.2 Control Strategies, Scenarios and Settings..........................50
2.9.3 Simulation Results..............................................................52
2.10 Conclusions..................................................................................56
References............................................................................................57
Adding Intelligence to Ubiquitous Computing Environments............61
Abstract.................................................................................................61
3.1 Introduction..................................................................................62
3.2 Ubiquitous Computing and Environments...................................64
3.2.1 Requirements Heterogeneity...............................................64
3.2.2 Background..........................................................................66
3.3 The intelligent Dormitory.............................................................69
3.4 The Intelligent Association System..............................................73
3.4.1 From Static Wiring to Ad hoc Networks.............................73
3.4.2 Embedded Agents and Societies.........................................74
3.4.3 Associations in IAS.............................................................77
3.5 The Fuzzy-IAS Agent..................................................................77
3.5.1 Designing Fuzzy Control Systems......................................78
3.5.2 The F-IAS Agent Model......................................................80
3.6 Converting Rules into Fuzzy Numerical Tables..........................85
3.7 Distribution with FuzzyXML ......................................................86
3.8 Online Intelligent Association Mechanisms.................................88
3.8.1 Why Intelligent Associations?.............................................88
3.8.2 Fuzzy Cognitive Maps (FCM)............................................89
3.8.3 Intelligent Association Calculation.....................................90
3.9 Experimental Results and Analysis..............................................93
3.9.1 Fuzzy Rule Extraction from Offline Collected Data...........95
3.9.2 Calculation of the Association Weights..............................97
3.10 Conclusions................................................................................100
Acknowledgement..............................................................................101
References..........................................................................................101
iJADE FreeWalker – An Intelligent Ontology Agent-based Tourist
Guiding System......................................................................................103
Abstract...............................................................................................103
4.1 Introduction................................................................................104