Behavior Computing: Modeling, Analysis, Mining and Decision PDF
Preview Behavior Computing: Modeling, Analysis, Mining and Decision
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
Description:The list of books you might like
