Behavior Computing: Modeling, Analysis, Mining and Decision

371 Pages·2012·7.05 MB·English

by  Nitin Agarwal

Behavior Computing
Longbing Cao (cid:2) Philip S. Yu
Editors
Behavior Computing
Modeling, Analysis, Mining and Decision
Editors
LongbingCao Prof.PhilipS.Yu
AdvancedAnalyticsInstitute DepartmentofComputerScience
UniversityofTechnologySydney UniversityofIllinoisatChicago
Sydney,NSW,Australia Chicago,IL,USA
ISBN978-1-4471-2968-4 ISBN978-1-4471-2969-1(eBook)
DOI10.1007/978-1-4471-2969-1
SpringerLondonHeidelbergNewYorkDordrecht
LibraryofCongressControlNumber:2012936583
©Springer-VerlagLondon2012
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
‘Behavior’isanincreasinglyimportantconceptinthescientific,societal,economic,
cultural, political, military, living and virtual world. In dictionaries, ‘Behavior’
refers to manner of behaving or acting, and the action or reaction of any mate-
rial under given circumstances. In Wikipedia, ‘behavior’ refers to the actions and
mannerismsmadebyorganisms,systemsorartificialentitiesinconjunctionwithits
environment,whichincludestheothersystemsororganismsaroundaswellasthe
physicalenvironment.Itistheresponseofthesystemororganismtovariousstimuli
orinputs,whetherinternalorexternal,consciousorsubconscious,overtorcovert,
andvoluntaryorinvoluntary.
Behaviorisubiquitous.Besidesthecommontermssuchasconsumerbehaviors,
humanbehaviors,animalbehaviors,andorganizationalbehaviors,behaviorsappear
everywhereatanytime.Behaviorsinthephysicalworldareexplicit,andhavebeen
studied from many different aspects. With the fast development and deep engage-
mentofsocialanddigitalizedlife,family,cityandplanetwithadvancedcomputing
technology, in particular, virtual reality, multimedia information processing, visu-
alization,machinelearning,patternrecognition,behaviorsinthevirtualandsocial
world are emerging increasingly. In addition, behaviors in the traditional spheres
arebecomingmoreandmorecomplexwiththeinvolvementandmarriagewiththe
virtualandsocialworld.
In different applications and scenarios, behaviors present respective character-
istics and features. For instance, in stock markets, trader’s behaviors are embod-
ied through trading actions and action properties, such as placing a buy quote at
a certain time, price and volume on a target security. The action, response or pre-
sentation associated with the corresponding properties forms a concrete and rich
object—behavior.
Therepresentation,modeling,analysis,datamininganddecision-makingofbe-
haviors are becoming increasingly useful, essential, and challenging in ubiquitous
behavioralapplicationsandproblem-solving.Theyformintoanewcomputingop-
portunity,necessityandtechnologyinnovation,werefertoitasbehaviorcomputing
orbehaviorinformatics.
v
vi Preface
Behavior computing, or behavior informatics, consists of methodologies, tech-
niquesandpracticaltoolsforrepresenting,modeling,analyzing,understandingand
utilizinghuman,organismal,organizational,societal,artificialandvirtualbehaviors,
behavioralinteractionsandrelationships,behavioralnetworks,behavioralpatterns,
behavioral impacts, the formation and decomposition of behavior-oriented groups
andcollectiveintelligence,andtheemergenceofbehavioralintelligence.Behavior
computing contributes to the in-depth understanding, discovery, applications and
managementofbehaviorintelligence.
TheaboveobservationsanddiscussionsmotivatetheeditingofthisbookBehav-
iorComputing:Modeling,Analysis,MiningandDecision.Theeditedbookreports
state-of-the-artadvancesinmethodologies,techniques,systemsandapplicationsof
behavior computing. Although there are some newly established conferences and
workshops, as well as special issues on behavior modeling and analysis of social
networks,thiseditedbookcreatesanimportantopportunitytobroadencurrentre-
searchtoareasthatconsistofbehaviors.Itaimstoserveasthefirstdedicatedsource
of references for the theory and applicationsof behaviorinformatics and behavior
computing, establishing state of the art research, disseminating the latest research
discoveries,andprovidingaground-breakingtextbooktoseniorundergraduateand
postgraduatestudents.
The book is composed of 23 chapters, which are selected from the 2010 and
2011 International Workshop on Behavior Informatics, submissions to this edited
book,partialsubmissionstotheSpecialIssueonBehaviorComputing,andinvited
chapters. The book consists of four parts, covering behavior modeling, behavior
analysis,behaviorminingandbehaviorapplications.
InPartI,thebookreportsattemptsandeffortsindevelopingrepresentationand
modeling methods and tools for capturing behavior characteristics and dynamics
inareassuchassocialmedia,soccergame,andsoftwarepackaging.Thisinvolves
newtechniquessuchasmodelinginfluentialbehaviorsinsocialmedia,abehavior
ontologysystemcalledSAPMASrepresentingsocialactivityprocess,usingnarra-
tiveknowledgerepresentationlanguagetorepresentbehaviors,andapplyingsemi-
Markovmodelstorepresentuserbehaviors.
Part II selects a number of the corresponding techniques for behavior analysis.
This involves great efforts to develop effective techniques and tools for emergent
areasanddomainsinanalyzingbehaviors,includingagroupbuyingbehaviorrec-
ommendationsystem,simultaneouslymodelingreplynetworksandcontentstogen-
erate user’s profiles on web forum, analyzing information searching behaviors by
reinforcement learning, estimating conceptual similarities by distributed represen-
tation and extended backpropagation, scoring and predicting risk preferences, and
creatingsimulatedfeedback.
Part III features behavior mining. The selected chapters address issues includ-
ing clustering trajectory routes, linking behavioral patterns to personal attributes,
miningcausalityfromnon-categoricalnumericaldata,mininghighutilityitemsets,
modelinganddetectionofsuspiciousactivities,abehavioralmodelingapproachto
discoverunauthorizedcopyingoflarge-scaledocuments,andanalyzingtwitteruser
behaviorsandtopictrends.
Preface vii
SixcasestudiesarereportedinPartIVonbehaviorapplications.Theycoverdo-
mainsandareasincludingtelecomuserbehaviors,eventdetectionincallingrecords,
predictingthenextcallforsmartphones,3Dhandwritingrecognitiononhandheld
devices,medicalstudentsearchbehaviors,andevaluationofsoftwaretestingstrate-
gies.
Theintendedaudienceofthisbookwillmainlyconsistofresearchers,research
studentsandpractitionersinbehaviorstudies,includinginthecommunitiesofcom-
puter science, behavioral science, and social science. In particular, this book fits
interests from behavior informatics, behavioral science, data mining, knowledge
representation,machinelearning,andknowledgediscovery.Thebookisalsoofin-
teresttoresearchersandindustrialpractitionersinareassuchasmarketinganalytics,
consumerbehavioranalysis,socialanalytics,onlinebehavioranalysis,businessan-
alytics, human-computer interaction, artificial intelligence, intelligent information
processing,decisionsupportsystems,andknowledgemanagement.
Readerswhoareinterestedinbehaviorcomputingandbehaviorinformaticsare
encouragedtorefertothespecialinterestgroup:BehaviorInformatics.TheSIGon
BehaviorInformaticsisadedicatedonlineresearchportalandrepository,presenting
researchoutcomesandopportunitiesontheoretical,technicalandpracticalissuesin
behaviorcomputingandbehaviorinformatics.
We would like to convey our appreciation to all contributors including the ac-
ceptedchapters’authors,andmanyotherparticipantswhosubmittedtheirchapters
thatcannotbeincludedinthebookduetospacelimits.OurspecialthankstoMs.
Helen Desmond and Mr. Ben Bishop from Springer UK and Ms. Melissa Fearon
from Springer US for their kind support and great efforts in bringing the book to
fruition. In addition, we also appreciate all reviewers, as well as Mr. Zhong She’s
assistanceinformattingthebook.
Sydney,Australia LongbingCao
Chicago,USA PhilipS.Yu
Contents
PartI BehaviorModeling
1 AnalyzingBehavioroftheInfluentialsAcrossSocialMedia . . . . . 3
NitinAgarwal,ShamanthKumar,HuijiGao,RezaZafarani,andHuanLiu
2 ModelingandAnalysisofSocialActivityProcess . . . . . . . . . . . 21
CanWangandLongbingCao
3 Behaviour Representation and Management Making Use
oftheNarrativeKnowledgeRepresentationLanguage . . . . . . . . 37
GianPieroZarri
4 Semi-MarkovianRepresentationofUserBehaviorinSoftware
Packages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
PrateetiMohapatraandHowardMichel
PartII BehaviorAnalysis
5 P-SERS:PersonalizedSocialEventRecommenderSystem . . . . . . 71
Yun-HuiHung,Jen-WeiHuang,andMing-SyanChen
6 Simultaneously Modeling Reply Networks and Contents
toGenerateUser’sProfilesonWebForum . . . . . . . . . . . . . . . 91
ZhaoZhang,WeiningQian,andAoyingZhou
7 InformationSearchingBehaviorMiningBasedonReinforcement
LearningModels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
LirenGan,YonghuaCen,andChenBai
8 Estimating Conceptual Similarities Using Distributed
RepresentationsandExtendedBackpropagation . . . . . . . . . . . 127
PeterDreisiger,WeiLiu,andCaraMacNish
9 ScoringandPredictingRiskPreferences . . . . . . . . . . . . . . . . 143
GürdalErtek,MuratKaya,CemreKefeli,ÖzgeOnur,andKeremUzer
ix
x Contents
10 AnIntroductiontoPrognosticSearch . . . . . . . . . . . . . . . . . 165
NithinKumarMandVasudevaVarma
PartIII BehaviorMining
11 Clustering Clues of Trajectories for Discovering Frequent
MovementBehaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Chih-ChiehHung,Ling-YinWei,andWen-ChihPeng
12 LinkingBehavioralPatternstoPersonalAttributesThroughData
Re-Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
GürdalErtek,AyhanDemiriz,andFatihCakmak
13 MiningCausalityfromNon-categoricalNumericalData . . . . . . . 215
JoaquimSilva,GabrielLopes,andAntónioFalcão
14 AFastAlgorithmforMiningHighUtilityItemsets . . . . . . . . . . 229
Show-JaneYen,Chia-ChingChen,andYue-ShiLee
15 IndividualMovementBehaviourinSecurePhysicalEnvironments:
ModelingandDetectionofSuspiciousActivity . . . . . . . . . . . . . 241
RobertP.Biuk-Aghai,Yain-WharSi,SimonFong,andPeng-FanYan
16 A Behavioral Modeling Approach to Prevent Unauthorized
Large-ScaleDocumentsCopyingfromDigitalLibraries . . . . . . . 255
EvgenyE.IvashkoandNataliaN.Nikitina
17 AnalyzingTwitterUserBehaviorsandTopicTrendsbyExploiting
DynamicRules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
LucaCaglieroandAlessandroFiori
PartIV BehaviorApplications
18 Behavior Analysis of Telecom Data Using Social Networks
Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
AvinashPolepallyandSaravananMohan
19 EventDetectionBasedonCallDetailRecords . . . . . . . . . . . . . 305
HuiqiZhangandRamDantu
20 SmartPhone:PredictingtheNextCall . . . . . . . . . . . . . . . . . 317
HuiqiZhangandRamDantu
21 ASystemwithHiddenMarkovModelsandGaussianMixture
Models for3D HandwritingRecognitiononHandheldDevices
UsingAccelerometers . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Wang-HsinHsu,Yi-YuanChiang,andJung-ShyrWu
22 MedicalStudents’SearchBehaviour:AnExploratorySurvey . . . . 337
AnushiaInthiran,SaadatM.Alhashmi,andPervaizK.Ahmed
Contents xi
23 AnEvaluationSchemeofSoftwareTestingStrategy . . . . . . . . . 353
K.AjayBabu,K.Madhuri,andM.Suman
AuthorIndex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
SubjectIndex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373