Semantic models for adaptive interactive systems

209 Pages·2013·6.255 MB·English

by  Hussein, Tim

Human–Computer Interaction Series
Editors-in-chief
JohnKarat
JeanVanderdonckt
UniversitécatholiquedeLouvain,Louvain-la-Neuve,Belgium
EditorialBoard
RavinBalakrishnan,UniversityofToronto,Toronto,ON,Canada
SimoneBarbosa,PUC-Rio,RiodeJaneiro,RJ,Brazil
ReginaBernhaupt,Ruwido,Salzburg,Austria
JohnCarroll,ThePennsylvaniaStateUniversity,UniversityPark,PA,USA
AdrianCheok,KeioUniversity,Tokyo,Japan
GilbertCockton,NorthumbriaUniversity,NewcastleuponTyne,UK
HenryBeen-LirnDuh,UniversityofTasmania,SandyBay,TAS,Australia
PeterForbrig,UniversitätRostock,Rostock,Germany
CarlaFreitas,FederalUniversityofRioGrandedoSul,PortoAlegre,RS,Brazil
HansGellersen,LancasterUniversity,Lancaster,UK
RobertJacob,TuftsUniversity,Medford,MA,USA
PanosMarkopoulos,EindhovenUniversityofTechnology,Eindhoven,TheNetherlands
GerritMeixner,HeilbronnUniversity,Heilbronn,Germany
DianneMurray,PuttingPeopleBeforeComputers,London,UK
BradA.Myers,CarnegieMellonUniversity,Pittsburgh,PA,USA
PhilippePalanque,UniversitéPaulSabatier,Toulouse,France
OscarPastor,UniversityofValencia,Valencia,Spain
BerylPlimmer,UniversityofAuckland,Auckland,NewZealand
DesneyTan,MicrosoftResearch,Redmond,WA,USA
ManfredTscheligi,CenterforUsabilityResearchandEngineering,Vienna,Austria
GerritvanderVeer,VrijeUniversiteitAmsterdam,Amsterdam,TheNetherlands
ShuminZhai,IBMAlmadenResearchCenter,SanJose,CA,USA
HCIisamultidisciplinaryfieldfocusedonhumanaspectsofthedevelopmentofcomputer
technology.Ascomputer-basedtechnologybecomesincreasinglypervasive—notjustinde-
velopedcountries,butworldwide—theneedtotakeahuman-centeredapproachinthedesign
anddevelopmentofthistechnologybecomesevermoreimportant.Forroughly30yearsnow,
researchersandpractitionersincomputationalandbehavioralscienceshaveworkedtoiden-
tifytheoryandpracticethatinfluencesthedirectionofthesetechnologies,andthisdiverse
work makes up the field of human-computer interaction. Broadly speaking it includes the
studyofwhattechnologymightbeabletodoforpeopleandhowpeoplemightinteractwith
thetechnology.TheHCIseriespublishesbooksthatadvancethescienceandtechnologyof
developingsystemswhicharebotheffectiveandsatisfyingforpeopleinawidevarietyof
contexts.Titlesfocusontheoreticalperspectives(suchasformalapproachesdrawnfroma
varietyofbehavioralsciences),practicalapproaches(suchasthetechniquesforeffectively
integratinguserneedsinsystemdevelopment),andsocialissues(suchasthedeterminantsof
utility,usabilityandacceptability).
Forfurthervolumes:
http://www.springer.com/series/6033
Tim Hussein (cid:2) Heiko Paulheim (cid:2)
Stephan Lukosch (cid:2) Jürgen Ziegler (cid:2) Gaëlle Calvary
Editors
Semantic Models
for Adaptive
Interactive Systems
Editors
TimHussein JürgenZiegler
InteractiveSystemsGroup InteractiveSystemsGroup
UniversityofDuisburg-Essen UniversityofDuisburg-Essen
Duisburg,Germany Duisburg,Germany
HeikoPaulheim GaëlleCalvary
DataandWebScienceGroup GrenobleInformaticsLaboratory
UniversityofMannheim GrenobleInstituteofTechnology
Mannheim,Germany Grenoble,France
StephanLukosch
FacultyofTechnology,Policy&Manageme
DelftUniversityofTechnology
Delft,TheNetherlands
ISSN1571-5035 Human–ComputerInteractionSeries
ISBN978-1-4471-5300-9 ISBN978-1-4471-5301-6(eBook)
DOI10.1007/978-1-4471-5301-6
SpringerLondonHeidelbergNewYorkDordrecht
LibraryofCongressControlNumber:2013943468
©Springer-VerlagLondon2013
Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpartof
thematerialisconcerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation,
broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,andtransmissionorinformation
storageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilarmethodology
nowknownorhereafterdeveloped.Exemptedfromthislegalreservationarebriefexcerptsinconnection
with reviews or scholarly analysis or material supplied specifically for the purpose of being entered
and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of
this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’slocation,initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.
PermissionsforusemaybeobtainedthroughRightsLinkattheCopyrightClearanceCenter.Violations
areliabletoprosecutionundertherespectiveCopyrightLaw.
Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication
doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant
protectivelawsandregulationsandthereforefreeforgeneraluse.
Whiletheadviceandinformationinthisbookarebelievedtobetrueandaccurateatthedateofpub-
lication,neithertheauthorsnortheeditorsnorthepublishercanacceptanylegalresponsibilityforany
errorsoromissionsthatmaybemade.Thepublishermakesnowarranty,expressorimplied,withrespect
tothematerialcontainedherein.
Printedonacid-freepaper
SpringerispartofSpringerScience+BusinessMedia(www.springer.com)
Preface
Semantic technologies and, in particular, ontologies as formal and shareable rep-
resentations of a domain play an increasingly important role in computer science,
especially for the design, development and execution of interactive systems. Se-
manticmodelscanserveanumberofdifferentpurposesinthiscontext.Theycanbe
used as functional core or user interface models in model-driven analysis, design,
generation,andadaptationofuserinterfaces.
Ontologiesmayenhancethefunctionalcoverageofaninteractivesystemaswell
as its visualization and interaction capabilities in various ways, e.g., by providing
inputassistance,intelligentlyclusteringinformation,guidingcollaborativeinterac-
tion, or adapting the user interface according to the user’s context. Especially in
thelattercase,ontologiescanbeappliedforrepresentingthevariouskindsofcon-
text information for context-aware and adaptive systems. In particular, they have
promised to provide a technique for representing external physical context factors
suchaslocation,timeortechnicalparameters,aswellas“internal”contextsuchas
userinterestprofilesorinteractioncontextinaconsistent,generalizedmanner.Ow-
ingtotheseproperties,semanticmodelscanalsocontributetobridginggaps,e.g.,
betweenusermodels,context-awareinterfacesandmodel-drivenUIgeneration.
Thereis,therefore,aconsiderablepotentialforusingsemanticmodelsasabasis
for adaptive interactive systems. The range of potential adaptations is wide com-
prising,forexample,context-anduser-dependentrecommendations,interactiveas-
sistance when performing application-specific tasks, adaptation of the application
functionality,adaptationofthecollaborationprocess,oradaptiveretrievalsupport.
Furthermore,avarietyofreasoningandmachinelearningtechniquesexist,thatcan
beemployedtoachieveadaptivesystembehavior.Last,butnotleast,theadventand
rapidgrowthofLinkedOpenDataasalarge-scalecollectionofsemanticdatahas
pavedthewayforanewbreedofintelligent,knowledge-intensiveapplications.
Toexplorethatpotential,wehaveestablishedaworkshopseriescalledSemantic
Models for Adaptive Interactive Systems (SEMAIS). The workshop had its debut
attheACMIntelligentUserInterfacesconferenceinHongKongin2010,andwas
followedbytwosubsequenteditionsinPaloAltoin2011,andinLisbonin2012.At
theworkshop,wehaveseencuttingedgeresearchspanningfromtheemploymentof
v
vi Preface
semanticmodelsinthedevelopmentandgenerationofinteractivesystemstonovel
interactionparadigmsandapplicationsforsemanticdata.
This book collects enhanced, revised, and updated versions of the best papers
submitted to the three workshops editions, as well as additional original contribu-
tions.Itprovidesinsightsintomethodologiesfordesigningadaptivesystemsbased
on semantic data, introduces models that can be used for building interactive sys-
tems,andshowcasesapplicationsmadepossiblebytheuseofsemanticmodels.
BookOutline
UI2Ont—A Formal Ontology on User Interfaces and Interactions by Heiko Paul-
heim and Florian Probst discusses the potentials of an encompassing ontology for
describinguserinterfacesandthewayhumansinteractwiththem.Theauthorsshow
how such an ontology can be constructed from existing user interface description
languagesanddescribehowitcanbeemployedforapplicationintegration.
GeneratingModelsofRecommendationProcessesoutofAnnotatedOntologies
by Hermann Kaindl et al. shows how the development of interactive systems—in
that case recommendation systems—can be automated to a certain extent by the
useofontologies.Theydiscussamethodologyforturningaproductontologyinto
a discourse system in which users can interactively choose products. The system
was tested in active online stores, showing that the semi-automatically generated
discourseswerecompetitivewithmanuallydesignedones.
CognitiveSemanticCategoriesasaBasisforaPrototypeAdaptiveInformation
System by Evangelos Kapros and Simon McGinnes introduces a methodology for
generating applicationsoffering basic general operations on a dynamic data struc-
ture.Theyleveragefindingsfromneurologyandcognitivesemanticstoderiveaset
of archetypal categories, which is used as a top level for automatically generating
intuitivevisualdesignsforadaptiveinformationsystems.
A Semantic Model for Adaptive Collaboration Support Systems by Stefan W.
Knolletal.discussesanencompassingframeworkforfosteringelasticcollaboration
processes,i.e.,collaborationprocessesthatarenotstaticallypredefined,butmaybe
adaptedtodynamicrequirementsandsituationalchanges.Theirapproachisbased
onasemanticmodelthatcanbeusedtoexpressinformationaboutprocessstepsas
wellastheparticipantsandtheircontexts,thusallowingfortheimplementationof
dynamicapplications.
A Semantics-Based, End-User-Centered Information Visualization Process for
Semantic Web Data by Martin Voigt et al. introduces the VizBoard workbench,
asystemwhichallowsenduserswithoutspecificSemanticWebskillstocreatein-
formativevisualizationsofSemanticWebdata.Byusingsemanticdescriptionofall
visualizationcomponents,complexadaptiveandinteractiveviewscanbegenerated.
PASTREM:ProactiveOntologyBasedRecommendationsforInformationWork-
ers by Benedikt Schmidt et al. addresses the needs of information workers deal-
ing with multiple diverse resources in various processes. The chapter discusses a
recommender system that detects the user’s current context and work process and
Preface vii
identifiesrelevantitemsintheuser’ssystem.Thesystemwasevaluatedusingdata
collected from different work stations at an IT company, and is shown to provide
moremeaningfulrecommendationsthancommonrecommendationalgorithms.
Visualizing Search Results of Linked Open Data by Christian Stab et al. in-
troduces an approach for making search on Linked Open Data more intuitive for
endusers.Theirapproachprovidesameanstotranslatenaturallanguagekeyword
searchestoformalqueriesonLinkedData,andgivestheusersvisualfeedbackon
boththesystem’sunderstandingoftheuser’squeryandthesearchresults.Theau-
thorsshowthatuserssearchingforinformationwiththeirsystemarebothfasteras
wellasmoresatisfiedthanwithtraditionalapproaches.
AContext-AwareShoppingPortalBasedonSemanticModelsbyTimHusseinet
al.illustrateshowsemanticmodelscanbeusedasbackenddatasourceforbothex-
plorationandadaptionofinteractivesystems.Theyshowhowsemanticmodelscan
beusedtoprovidefacetedbrowsingaswellasuseradaptionandrecommendation,
usingspreadingactivitationonsemanticdatatomakethesystemadapttoauser’s
preferences.
SemanticModelsforInteractiveSystems:TheCaseofTaggingandFolksonomies
bySteffenLohmannisconcernedwithaspecificinteractiontechniquethathasbe-
come popular in the Web 2.0, i.e., tagging. User generated tags are used as a ba-
sis for finding and recommending content in large-scale platforms such as Flickr
or YouTube. The chapter introduces a formal ontology for describing tagging in-
teractions and the relations between individual tags, which can be used for novel
graphicalvisualizations.
UserInteractionTemplatesfortheDesignofLifeloggingSystemsbyFrankHopf-
gartneretal.showshowsemanticscanhelporganizingandanalyzingtheabundance
of data generated by lifelogging systems, i.e., systems that constantly track their
users.Theydiscussusecases,interactiontechniques,andinformationvisualization
approaches that are made possible by using semantic representations of the data
collectedbylifeloggingsystems.
Acknowledgements
Theeditorswouldliketothankalltheauthorscontributingchapterstothisvolume
and participating in the peer reviewing process, as well as Beverley Ford and Ben
BishopatSpringerformakingthisvolumepossible.Furthermore,wewouldliketo
thanktheorganizersoftheACMIntelligentUserInterfacesconferences2010–2012
forprovidingtheframefortheSEMAISworkshopseries,andalltheparticipantsat
thoseworkshopsfortheirinterestingsubmissionsandstimulatingdiscussions.
Duisburg,Mannheim,Delft,Grenoble TimHussein
March2013 HeikoPaulheim
StephanLukosch
JürgenZiegler
GaëlleCalvary
Contents
1 UI2Ont—AFormalOntologyonUserInterfacesandInteractions . . 1
HeikoPaulheimandFlorianProbst
2 GeneratingModelsofRecommendationProcessesoutofAnnotated
Ontologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
HermannKaindl,DominikErtl,RomanPopp,RalphHoch,JürgenFalb,
EdinArnautovic,AdaOkoli,andMartinSchliefnig
3 CognitiveSemanticCategoriesasaBasisforaPrototypeAdaptive
InformationSystem . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
EvangelosKaprosandSimonMcGinnes
4 ASemanticModelforAdaptiveCollaborationSupportSystems . . . 59
StefanW.Knoll, JordanJaneiro, StephanG.Lukosch, and
GwendolynL.Kolfschoten
5 ASemantics-Based,End-User-CenteredInformationVisualization
ProcessforSemanticWebData . . . . . . . . . . . . . . . . . . . . . 83
MartinVoigt,StefanPietschmann,andKlausMeißner
6 PASTREM: Proactive Ontology Based Recommendations for
InformationWorkers . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
BenediktSchmidt,EickeGodehardt,andHeikoPaulheim
7 VisualizingSearchResultsofLinkedOpenData . . . . . . . . . . . 133
ChristianStab,DirkBurkhardt,MatthiasBreyer,andKawaNazemi
8 AContext-AwareShoppingPortalBasedonSemanticModels . . . . 151
TimHussein,TimmLinder,andJürgenZiegler
9 SemanticModelsforInteractiveSystems:TheCaseofTaggingand
Folksonomies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
SteffenLohmann
10 UserInteractionTemplatesfortheDesignofLifeloggingSystems . . 187
FrankHopfgartner,YangYang,LijuanMarissaZhou,andCathalGurrin
ix
Contributors
EdinArnautovic Vienna,Austria
Matthias Breyer Fraunhofer Institute for Computer Graphics Research (IGD),
Darmstadt,Germany
Dirk Burkhardt Fraunhofer Institute for Computer Graphics Research (IGD),
Darmstadt,Germany
DominikErtl ViennaUniversityofTechnology,Vienna,Austria
JürgenFalb ViennaUniversityofTechnology,Vienna,Austria
EickeGodehardt SAPResearch,Darmstadt,Germany
CathalGurrin DublinCityUniversity,Dublin,Ireland
RalphHoch ViennaUniversityofTechnology,Vienna,Austria
FrankHopfgartner TUBerlin,Berlin,Germany
TimHussein UniversityofDuisburg-Essen,Duisburg,Germany
JordanJaneiro DelftUniversityofTechnology,Delft,TheNetherlands
HermannKaindl ViennaUniversityofTechnology,Vienna,Austria
EvangelosKapros TheUniversityofDublin,Dublin,Ireland
StefanW.Knoll DelftUniversityofTechnology,Delft,TheNetherlands
GwendolynL.Kolfschoten DelftUniversityofTechnology,Delft,
TheNetherlands
TimmLinder UniversityofDuisburg-Essen,Duisburg,Germany
SteffenLohmann UniversityofStuttgart,Stuttgart,Germany
StephanG.Lukosch DelftUniversityofTechnology,Delft,TheNetherlands
SimonMcGinnes TheUniversityofDublin,Dublin,Ireland
xi