Transactions on Computational Collective Intelligence II

203 Pages·2010·4.515 MB·English

by  Grzegorz J. Nalepa, Weronika T. Furmańska (auth.), Ngoc Thanh Nguyen, Ryszard Kowalczyk (eds.)

Lecture Notes in Computer Science 6450
CommencedPublicationin1973
FoundingandFormerSeriesEditors:
GerhardGoos,JurisHartmanis,andJanvanLeeuwen
EditorialBoard
DavidHutchison
LancasterUniversity,UK
TakeoKanade
CarnegieMellonUniversity,Pittsburgh,PA,USA
JosefKittler
UniversityofSurrey,Guildford,UK
JonM.Kleinberg
CornellUniversity,Ithaca,NY,USA
AlfredKobsa
UniversityofCalifornia,Irvine,CA,USA
FriedemannMattern
ETHZurich,Switzerland
JohnC.Mitchell
StanfordUniversity,CA,USA
MoniNaor
WeizmannInstituteofScience,Rehovot,Israel
OscarNierstrasz
UniversityofBern,Switzerland
C.PanduRangan
IndianInstituteofTechnology,Madras,India
BernhardSteffen
TUDortmundUniversity,Germany
MadhuSudan
MicrosoftResearch,Cambridge,MA,USA
DemetriTerzopoulos
UniversityofCalifornia,LosAngeles,CA,USA
DougTygar
UniversityofCalifornia,Berkeley,CA,USA
GerhardWeikum
MaxPlanckInstituteforInformatics,Saarbruecken,Germany
Ngoc Thanh Nguyen
Ryszard Kowalczyk (Eds.)
Transactions on
Computational
Collective Intelligence II
1 3
VolumeEditors
NgocThanhNguyen
WroclawUniversityofTechnology
InstituteofInformatics
Str.Wyb.Wyspianskiego27
50-370Wroclaw,Poland
E-mail:[e-mail]
RyszardKowalczyk
SwinburneUniversityofTechnology
CentreforComplexSoftwareSystemsandServices
P.O.Box218
Hawthorn,Victoria3122,Australia
E-mail:[e-mail]
LibraryofCongressControlNumber:Appliedfor
CRSubjectClassification(1998):I.2,C.2.4,I.2.11,H.3-5,D.2
ISSN 0302-9743
ISBN-10 3-642-17154-0SpringerBerlinHeidelbergNewYork
ISBN-13 978-3-642-17154-3SpringerBerlinHeidelbergNewYork
Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis
concerned,specificallytherightsoftranslation,reprinting,re-useofillustrations,recitation,broadcasting,
reproductiononmicrofilmsorinanyotherway,andstorageindatabanks.Duplicationofthispublication
orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,1965,
initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violationsareliable
toprosecutionundertheGermanCopyrightLaw.
springer.com
©Springer-VerlagBerlinHeidelberg2010
PrintedinGermany
Typesetting:Camera-readybyauthor,dataconversionbyScientificPublishingServices,Chennai,India
Printedonacid-freepaper 06/3180
Preface
Welcome to the second volume of Transactions on Computational Collective Intelli-
gence (TCCI), a new journal devoted to research in computer-based methods of com-
putational collective intelligence (CCI) and their applications in a wide range of fields
such as the Semantic Web, social networks and multi-agent systems. TCCI strives to
cover new methodological, theoretical and practical aspects of CCI understood as the
form of intelligence that emerges from the collaboration and competition of many
individuals (artificial and/or natural). The application of multiple computational intel-
ligence technologies such as fuzzy systems, evolutionary computation, neural sys-
tems, consensus theory, etc., aims to support human and other collective intelligence
and to create new forms of CCI in natural and/or artificial systems.
TCCI is a double-blind refereed and authoritative reference dealing with the work-
ing potential of CCI methodologies and applications, as well as emerging issues of
interest to academics and practitioners. This second issue contains a collection of 10
articles selected from high-quality submissions addressing advances in the founda-
tions and applications of computational collective intelligence. In “Integration Pro-
posal for Description Logic and Attributive Logic – Towards Semantic Web Rules”
G. Nalepa and W. Furmanska propose a transition from attributive logic to description
logic in order to improve the design of Semantic Web rules. K. Thorisson et al. in
“The Semantic Web: From Representation to Realization” present key ideas behind
SemCard technology and its initial implementation, aiming to support management of
the full lifecycle of data in the Semantic Web. “A Cross-cultural Multi-agent Model
of Opportunism in Trade” by G. Hofstede et al. presents a model of deceit and trust in
trade, together with a comparative multi-agent simulation of trading situations.
L. Longo et al. in “Enhancing Social Search: A Computational Collective Intelligence
Model of Behavioral Trait, Trust and Time” describe a social search model based on
information foraging theory, effort and computational trust, showing a different way
to implicitly judge Web-entities. “Group-Oriented Services: A Shift Toward Con-
sumer-Managed Relationships in the Telecom Industry” by L. Vrdoljak et al. intro-
duces an idea of a special type of personalized telecom services, called group-oriented
services, and its application in agent-based mobile content brokerage. “Pricing the
Services in Dynamic Environment: Agent Pricing Model” by D. Zagar et al. proposes
an agent-based pricing architecture to enable objective and transparent assessment of
the cost of the services. The next paper in this issue entitled “A Robust Approach for
Nonlinear UAV Task Assignment Problem Under Uncertainty” by H.A. Le and Q.T.
Nguyen includes a new robust approach to the task assignment of unmanned aerial
vehicles operating in uncertain environments whose objective is maximizing the tar-
get score. In “Decision Support System Based on Computational Collective Intelli-
gence in Campus Information Systems” Y. Saito and T. Matsuo describe a method
and an application of reusing campus collective information. “Fuel Crime Conceptu-
alization Through Specialization of Ontology for Investigation Management System”
by J. Cybulka presents a conceptual model based on consensual semantics to support
VI Preface
the teamwork of investigators of economics crimes. Finally, D. Barbucha et al. in
“JABAT Middleware as a Tool for Solving Optimization Problems” present experi-
ence in developing several applications of JABAT, a middleware supporting A-Team
architecture for solving optimization problems.
The research area of CCI has been growing significantly in recent years and we are
very thankful to everyone within the CCI research community who has supported
the Transactions on Computational Collective Intelligence and its affiliated events
including the International Conferences on Computational Collective Intelligence:
Semantic Web, Social Networks & Multiagent Systems (ICCCI). With this strong
support and a large number of submissions we are very pleased that TCCI and ICCCI
are being cemented as high-quality platforms for presenting and exchanging the most
important and significant advances in CCI research and development. We would like
to thank all the authors for their contributions to TCCI. This issue would also not have
been possible without the great efforts of the editorial board and many anonymously
acting reviewers. Here, we would like to express our sincere thanks to all of
them. Finally, we would also like to express our gratitude to the LNCS editorial
staff of Springer, in particular Alfred Hofmann, Ursula Barth and their team, who
have supported the TCCI journal and the editorship of this issue in a very
professional way.
September 2010 Ngoc Thanh Nguyen
Ryszard Kowalczyk
Transactions on Computational Collective Intelligence
This new journal focuses on research in applications of the computer-based methods
of computational collective intelligence (CCI) and their applications in a wide range
of fields such as the Semantic Web, social networks and multi-agent systems. It aims
to provide a forum for the presentation of scientific research and technological
achievements accomplished by the international community.
The topics addressed by this journal include all solutions of real-life problems for
which it is necessary to use computational collective intelligence technologies to achieve
effective results. The emphasis of the papers published is on novel and original research
and technological advancements. Special features on specific topics are welcome.
Editors-in-Chief
Ngoc Thanh Nguyen Wroclaw University of Technology, Poland
Ryszard Kowalczyk Swinburne University of Technology, Australia
Editorial Board
John Breslin National University of Ireland, Galway, Ireland
Shi-Kuo Chang University of Pittsburgh, USA
Oscar Cordon European Centre for Soft Computing, Spain
Tzung-Pei Hong National University of Kaohsiung, Taiwan
Gordan Jezic University of Zagreb, Croatia
Piotr J(cid:266)drzejowicz Gdynia Maritime University, Poland
Kang-Huyn Jo University of Ulsan, Korea
Radosław Katarzyniak Wroclaw University of Technology, Poland
Jozef Korbicz University of Zielona Gora, Poland
Hoai An Le Thi Metz University, France
Pierre Lévy University of Ottawa, Canada
Tokuro Matsuo Yamagata University, Japan
Kazumi Nakamatsu University of Hyogo, Japan
Toyoaki Nishida Kyoto University, Japan
Manuel Núñez Universidad Complutense de Madrid, Spain
Julian Padget University of Bath, UK
Witold Pedrycz University of Alberta, Canada
Debbie Richards Macquarie University, Australia
Roman Słowi(cid:276)ski Poznan University of Technology, Poland
Edward Szczerbicki University of Newcastle, Australia
Kristinn R. Thorisson Reykjavik University, Iceland
Gloria Phillips-Wren Loyola University Maryland, USA
Sławomir Zadro(cid:298)ny Institute of Research Systems, PAS, Poland
Table of Contents
Integration Proposal for Description Logic and Attributive
Logic – Towards Semantic Web Rules............................... 1
Grzegorz J. Nalepa and Weronika T. Furman´ska
A Cross-CulturalMulti-agent Model of Opportunism in Trade......... 24
Gert Jan Hofstede, Catholijn M. Jonker, and Tim Verwaart
Enhancing Social Search: A Computational Collective Intelligence
Model of BehaviouralTraits, Trust and Time........................ 46
Luca Longo, Pierpaolo Dondio, and Stephen Barrett
Group-Oriented Services: A Shift towards Consumer-Managed
Relationships in the Telecom Industry .............................. 70
Luka Vrdoljak, Iva Bojic, Vedran Podobnik, Gordan Jezic, and
Mario Kusek
The Semantic Web: From Representation to Realization .............. 90
Kristinn R. Tho´risson, Nova Spivack, and James M. Wissner
Decision Support System Based on Computational Collective
Intelligence in Campus Information Systems......................... 108
Yoshihito Saito and Tokuro Matsuo
Fuel Crime Conceptualization through Specialization of Ontology for
Investigation Management System ................................. 123
Jolanta Cybulka
A Robust Approach for Nonlinear UAV Task Assignment Problem
under Uncertainty ............................................... 147
Le Thi Hoai An and Nguyen Quang Thuan
Pricing the Services in Dynamic Environment: Agent Pricing Model .... 160
Drago Zˇagar, Slavko Rupˇci´c, and Snjeˇzana Rimac-Drlje
JABAT Middleware as a Tool for Solving Optimization Problems ...... 181
Dariusz Barbucha, Ireneusz Czarnowski, Piotr Je¸drzejowicz,
Ewa Ratajczak-Ropel, and Izabela Wierzbowska
Author Index.................................................. 197
Integration Proposal for Description Logic and
(cid:2)
Attributive Logic – Towards Semantic Web Rules
Grzegorz J. Nalepa and Weronika T. Furmańska
Instituteof Automatics,
AGHUniversity of Science and Technology,
Al. Mickiewicza 30, 30-059 Kraków, Poland
[e-mail], [e-mail]
Abstract. The current challenge of the Semantic Web is the develop-
mentofanexpressiveyeteffectiverulelanguage.Thispaperpresentsan
integration proposal for Description Logics (DL)and AttributiveLogics
(ALSV) is presented. These two formalisms stem from fields of Knowl-
edgeRepresentationandArtificialIntelligence.However,theyarebased
on different design goals and therefore provide different description and
reasoning capabilities. ALSV is the foundation of XTT2, an expressive
language for rule-based systems. DL provide formulation for expressive
ontology languages such as OWL2. An important research direction is
thedevelopmentofrulelanguagesthatcanbeintegratedwithontologies.
The contribution of the paper consists in introducing a possible transi-
tion from ALSV to DL. This opens up possibilities of using XTT2, a
well-founded rule-based system modelling rule language, to improvethe
design of SemanticWeb rules1.
1 Introduction
The Semantic Webproposal[2],ofthe next generationWeb withrichsemantics
and automated inference is based on a number of formalconcepts and practical
technologies. The former includes Description Logics (DL) [3] as the formalism
fordescribingontologies.Currently,theSemanticWebdevelopmentisfocusedon
providingaflexiblerulelanguagefortheWeb.ItshouldbeRIF-compatible[4]on
theruleinterchangelevel,andconceptuallycompatiblewithontologiesmodelled
in OWL with the use of Description Logics.Among other proposals,the SWRL
language [5] aims at meeting these requirements.
The Semantic Web initiativeis basedonpreviousexperiencesandresearchof
Knowledge Engineering [6] in the field of Artificial Intelligence [7]. In this field
rule-basedexpertsystemstechnologiesareaprimeexampleofeffectivereasoning
systems based on the rule paradigm [8,9,10]. The formal description of these
(cid:2) The paper is supported by the BIMLOQ Project funded from 2010–2012 resources
for science as a research project.
1 The paper extends concepts and preliminary results described in the paper [1] pre-
sented at ICCCI’09 Conference in Wrocław, Poland.
N.T.NguyenandR.Kowalczyk(Eds.):TransactionsonCCIII,LNCS6450,pp.1–23,2010.
(cid:2)c Springer-VerlagBerlinHeidelberg2010
2 G.J. Nalepa and W.T. Furmańska
systems is based on the propositional calculus, or restricted form of predicate
logic – like in the case of Prolog [11]. It is worth noting how the Semantic Web
community could benefit from classic rule-based systems’ tools and solutions.
A recent proposal of a new logical calculus extends the expressiveness of
rulelanguagesbyintroducinganattributivelanguageforruleformulation[10,12].
This solution seems superior to the simple propositional systems, and still
easier to reasonwith than the classicalpredicate logic.XTT2 is a rule language
based on this approach [12,13]. It provides visual design and formal analysis
methodsfordecisionrules.Thesefeatureswouldbe beneficialforSemanticWeb
applications.However,thecurrentproblemwithAttributiveLogicwithSetVal-
uesoverFiniteDomain(ALSV(FD))isthelackofconceptualcompatibilitywith
DL.
The rest of this paper is organised as follows: selected DL concepts are
discussed in Sect. 2, and a brief introduction to ALSV(FD) is given in Sect. 3.
Sect. 4 givesa motivationfor the researchaiming attranslationfromthe ALSV
(FD) to DL. ALSV(FD) is oriented at forward chaining rule-based systems,
while DL providesa formalizedfoundationfor ontologies.Syntaxand semantics
of these two calculi are compared in Sect. 5. In Sect. 6 an integration proposal
togetherwith a descriptionlanguageis introduced.A simple example transition
is then presented in Sect. 7. In Sect. 8 the proposal is evaluated and compared
with selected existing solutions in Sect. 9. The paper ends with ideas for future
work.
2 Description Logics Overview
DescriptionLogicsareafamilyofknowledgerepresentationlanguages[3].Histor-
icallyrelatedtosemanticnetworksandframelanguages,theydescribetheworld
ofinterestbymeansofconcepts,individualsandroles.However,contrarytotheir
predecessors such as semantic networks [6], they provide formal semantics and
thus allow for automated reasoning.Basic Description Logics take advantage of
theirrelationtopredicatecalculus.Ononehandtheyadoptitssemantics,which
makesthemmoreexpressivethanthepropositionallogic.Ontheotherhand,by
restricting the syntax to formulae with at most two variables, they remain de-
cidableandmorehuman-readable.ThesefeatureshavemadeDescriptionLogics
a popular formalism used for designing ontologies for the Semantic Web. There
exists a number of DL languages that are defined and distinguished by which
concept descriptions they allow, which influences the languages’ expressivity.
The vocabulary in DL consists of languages are concepts, which denote sets
ofindividuals, and roles, whichdenote the binaryrelations betweenindividuals.
ElementarydescriptionsinDLareatomic concepts andatomic roles.Morecom-
plexdescriptionscanbe builtinductivelyfromthoseusingconcept constructors.
Respective DL languages are distinguished by the constructors they provide. A
minimal language of practical interest is the Attributive Language [3] (AL).
Integration Proposal for Description Logic 3
Some of the most important definitions are presented (see [3]) below.
Definition 1. Let A denote an atomic concept and R an atomic role. In basic
AL concept descriptions C and D can be in one of the following forms:
A atomic concept (1)
(cid:2) universal concept (2)
⊥ bottom concept (3)
¬A atomic negation (4)
C(cid:4)D intersection (5)
∀R.C value restriction (6)
∃R.(cid:2) limited existential quantification (7)
In order to define a formal semantics, an interpretation I = (ΔI,·I) is con-
sidered. This interpretation consists of the domain of interpretation which is
a non-empty set and an interpretation function, which to every atomic con-
cept A assigns a set AI ⊆ ΔI and for every atomic role R a binary relation
RI = RI ⊆ ΔI × ΔI. The interpretation function is extended over concept
descriptions by Definition 2.
Definition 2.
(cid:2)I =ΔI (8)
⊥I =∅ (9)
(¬A)I =ΔI \AI (10)
(C(cid:4)D)I =CI ∩DI (11)
(∀R.C)I ={a∈ΔI|∀b,(a,b)∈RI →b∈CI} (12)
(∃R.(cid:2))I ={a∈ΔI|∃b,(a,b)∈RI} (13)
Thebasiclanguagecanbeextendedbyallowingotherconceptconstructors,such
as union (U) , full negation (C), full existential quantification (E) or number
restriction (N). Resulting formalisms are called using the letters indicating the
allowedconstructors,e.g.ALC,ALCN,ALUE etc.Thesmallestpropositionally
closed language is ALC [3].
Basic DL allowonly atomic roles (i.e. rolenames) in roledescriptions.Differ-
ent extensions are introduced by allowing role constructors. They enable intro-
duction of various constraints and properties of roles,such as transitive closure,
intersection, composition and union, or complement and inverse roles. Another
kind of extension is obtained by allowing nominals in concept definitions and
introducing primitive datatypes (see [3]). These modifications proved to be ex-
tremelyvaluableandimportantinthecontextoftheSemanticWebandontology
engineering.However,they aresourcesofhighcomputationalcomplexityofrea-
soning in the resulting ontologies.