Trends in Social Network Analysis: Information Propagation, User Behavior Modeling, Forecasting, and Vulnerability Assessment PDF
Preview Trends in Social Network Analysis: Information Propagation, User Behavior Modeling, Forecasting, and Vulnerability Assessment
Lecture Notes in Social Networks Rokia Missaoui Talel Abdessalem Matthieu Latapy Editors Trends in Social Network Analysis Information Propagation, User Behavior Modeling, Forecasting, and Vulnerability Assessment Lecture Notes in Social Networks SeriesEditors RedaAlhajj,UniversityofCalgary,Calgary,AB,Canada UweGlässer,SimonFraserUniversity,Burnaby,BC,Canada AdvisoryBoard CharuC.Aggarwal,IBMT.J.WatsonResearchCenter,Hawthorne,NY,USA PatriciaL.Brantingham,SimonFraserUniversity,Burnaby,BC,Canada ThiloGross,UniversityofBristol,Bristol,UK JiaweiHan,UniversityofIllinoisatUrbana-Champaign,IL,USA HuanLiu,ArizonaStateUniversity,Tempe,AZ,USA RaulManasevich,UniversityofChile,Santiago,Chile AnthonyJ.Masys,CentreforSecurityScience,Ottawa,ON,Canada CarloMorselli,UniversityofMontreal,QC,Canada RafaelWittek,UniversityofGroningen,TheNetherlands DanielZeng,TheUniversityofArizona,Tucson,AZ,USA Moreinformationaboutthisseriesathttp://www.springer.com/series/8768 Rokia Missaoui (cid:129) Talel Abdessalem Matthieu Latapy Editors Trends in Social Network Analysis Information Propagation, User Behavior Modeling, Forecasting, and Vulnerability Assessment 123 Editors RokiaMissaoui TalelAbdessalem DepartmentofComputerScience DepartmentofComputerScience &Engineering andNetworks UniversityofQuebecinOutaouais TelecomParisTech Gatineau,QC,Canada Paris,France MatthieuLatapy UPMCUnivParis06,CNRS LIP6UMR7606 SorbonneUniversitès Paris,France ISSN2190-5428 ISSN2190-5436 (electronic) LectureNotesinSocialNetworks ISBN978-3-319-53419-0 ISBN978-3-319-53420-6 (eBook) DOI10.1007/978-3-319-53420-6 LibraryofCongressControlNumber:2017936483 ©SpringerInternationalPublishingAG2017 Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpartof thematerialisconcerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation, broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,andtransmissionorinformation storageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilarmethodology nowknownorhereafterdeveloped. Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant protectivelawsandregulationsandthereforefreeforgeneraluse. Thepublisher,theauthorsandtheeditorsaresafetoassumethattheadviceandinformationinthisbook arebelievedtobetrueandaccurateatthedateofpublication.Neitherthepublishernortheauthorsor theeditorsgiveawarranty,expressorimplied,withrespecttothematerialcontainedhereinorforany errorsoromissionsthatmayhavebeenmade.Thepublisherremainsneutralwithregardtojurisdictional claimsinpublishedmapsandinstitutionalaffiliations. Printedonacid-freepaper ThisSpringerimprintispublishedbySpringerNature TheregisteredcompanyisSpringerInternationalPublishingAG Theregisteredcompanyaddressis:Gewerbestrasse11,6330Cham,Switzerland Preface A rapidly growing number of Internet users participate in social networks, belong to communities, produce, broadcast and use media content in different ways. This tremendous growth of social networks has led to a multidisciplinary research on analyzing and mining such networks according to various directions and trends as showninthepresentbook. The book contains recent studies in social network analysis and represent extended versions of a selected collection of articles presented at the 2015 IEEE/ACM international Conference on Advances in Social Network Analysis and Mining (ASONAM), which took place in Paris between August 25 and 28, 2015. The topics covered by this book are: assortativity, influence propagation andmaximization,socialuser’sprofileandbehavioralmodeling,sarcasmanalysis, delurking, social engineering and vulnerability assessment, link prediction and socialmediaforecasting. “ThePerceivedAssortativityofSocialNetworks:MethodologicalProblemsand Solutions” by David N. Fisher, Matthew J. Silk and Daniel W. Franks studies assortativity(alsocalledassortativemixing)insocialnetworksandshowsthatwhile socialnetworksaremoreassortativethannon-socialones,theytendtobepositively assortativeonlywhentheyarebuiltusinggroup-basedmethods.Toovercomethis bias,anumberofsolutionsbasedonadvancesinsociologicalandbiologicalfields areexploited. “A Parametric Study to Construct Time-aware Social Profiles” by Sirinya On- at,ArnaudQuirin,AndréPéninou,NadineBaptiste-Jessel,Marie-FrançoiseCanut, FlorenceSèdesfocusesonusinginformationsharedonusers’egocentricnetworks to extract their interests. To that end, a time-aware method is applied inside an existingsocialprofilebuildingprocessandaimsatweightinguser’sinterestsinthe socialprofileaccordingtotheirtemporalrelevance(temporalscore).Anempirical studyonscientificpublicationnetworks(DBLP/Mendeley)isthenusedtocompare theeffectivenessandaccuracyoftheproposedsocialprofileconstructiontechnique againstthetimeagnostictechnique. v vi Preface “Sarcasm Analysis on Twitter Data Using Machine Learning Approaches” by SantoshKumarBharti,RamkrushnaPradhan,KorraSathyaBabuandSanjayKumar Jenatacklesthenewandchallengingproblemofsarcasmanalysisinsocialnetworks by proposing four approaches that aim at extracting text features such as lexical, hyperbole, behavioral and universal facts. Then, the accuracy of the proposed solutionsforTwitterdataiscomputedandrevealsaconsiderableimprovementover existingsarcasmidentificationtechniques. The goal of the chapter, “The DEvOTION algorithm for Delurking in Social Networks”byRobertoInterdonato,ChiaraPuliceandAndreaTagarelliistodelurk silentmembersofasocialnetwork,i.e.,toencouragethemtogetmoreinvolvedin thenetwork.Toreachsuchagoal,theauthorsdefineadelurking-orientedtargeted influence maximization problem under the linear threshold model and propose an approximate solution based on a greedy algorithm named DEvOTION. The superiorityofthedefinedprocedureoverotherexistingdelurkingapproachesisthen given. “Social Engineering Threat Assessment using a Multi-layered Graph-based Model” by Omar Jaafor and Babiga Birregah deals with a security issue related to social engineering vulnerability, i.e., evaluating the set of attacks that focus on deceiving humans into performing actions or disclosing information. The study proposes a graph-based multi-layered model builtfromlayers that depict different attack scenarios, different states in an attack and multiple contexts that could be used.Ithighlightstheinterconnectionsbetweenthedifferentelementsinanattack suchasactionsperformed,usersinvolvedandresourcesused,andrepresentsactions thatdonotnecessarilyleavetracesinamonitoredsystem. In the chapter titled “Through The Grapevine: A Comparison of News in MicroblogsandTraditionalMedia”,theauthorsByungkyuKang,HaleighWright, Tobias Höllerer, Ambuj K. Singh and John O’Donovan propose two novel algo- rithmic approaches, namely content similarity computation and graph analysis, to automatically capture the main differences in newsworthy content between microblogsandtraditionalnewsmedia. “PredictionofElevatedActivityinOnlineSocialMediaUsingAggregatedand Individualized Models” by Jimpei Harada, David Darmon, Michelle Girvan and William Rand deals with information propagation to a large set of social network membersbyfocusingontheidentificationofthetimeperiodswhenalargeportion of a target population is active, which requires modeling users’ behavior. Three methods for behavior modeling are then proposed and validated on data collected fromasetofusersonTwitterin2011and2012. “Unsupervised Link Prediction based on time frames in Weighted-Directed CitationNetworks”byMehmetKaya,MujtabaJawed,ErtanBütünandRedaAlhajj proposesatime-framebasedunsupervisedlinkpredictionmethodfordirectedand weightednetworks.Tothatend,weightedtemporaleventsarefirstdefined.Then,a novelapproachbasedonthecommonneighbormetricforcomputingthetime-frame based node score is given. The empirical study exploits an unsupervised learning strategyonaweighted-directedcitationnetworktoshowthattheproposedmethod givesaccuratepredictionandpromisingresults. Preface vii “An Approach to Maximize the Influence Spread in Social Networks” by Ibrahima Gaye, Gervais Mendy, Samuel Ouya and Djaraf Seck deals with the influence maximization problem by proposing a Spanning Connected Graph algo- rithm (with three variants) that computes the seeds from which the information propagationisinitiated.Thefirstvariantbuildsthechildrenofthenodesrandomly whilethesecondoneusestheneighborhoodfortheidentificationofchildren.The third variant is a generalization of the first two ones and takes an arbitrary graph as input while the first two variants require a connected graph as input. These proceduresareeffectiveandhaveapolynomialtimecomplexity. Eva García Martín, Niklas Lavesson, and Håkan Grahn on “Energy Efficiency AnalysisoftheVeryFastDecisionTreeAlgorithm”addressesageneralissueindata mining applications and could be useful for mining social networks. It introduces energy consumption and energy efficiency as important factors to consider during dataminingalgorithmanalysisandevaluation.Theimpactofvaryingtheparameters of the Very Fast Decision Tree (VFDT) algorithm on energy consumption and accuracy is empirically studied. The conclusion is that energy consumption is affected by such parameters and can be reduced significantly while maintaining accuracy. We would like to conclude this preface by conveying our appreciation to all contributing authors and our warm thanks to Professor Reda Alhajj for giving us theopportunitytobetheGuestEditorsofthisbook.Wealsowouldliketoexpress ourgratitudetoChristopherT.CoughlinandhisteammembersfromSpringerUS fortheirhelpinthepreparationofthisvolume. QC,Canada RokiaMissaoui Paris,France TalelAbdessalem Paris,France MatthieuLatapy December2016 Contents ThePerceivedAssortativityofSocialNetworks:Methodological ProblemsandSolutions.......................................................... 1 DavidN.Fisher,MatthewJ.Silk,andDanielW.Franks AParametricStudytoConstructTime-AwareSocialProfiles.............. 21 SirinyaOn-at,ArnaudQuirin,AndréPéninou,NadineBaptiste-Jessel, Marie-FrançoiseCanut,andFlorenceSèdes SarcasmAnalysisonTwitterDataUsingMachineLearning Approaches ....................................................................... 51 SantoshKumarBharti,RamkrushnaPradhan,KorraSathyaBabu, andSanjayKumarJena TheDEvOTIONAlgorithmforDelurkinginSocialNetworks............. 77 RobertoInterdonato,ChiaraPulice,andAndreaTagarelli Social Engineering Threat Assessment Using a Multi-Layered Graph-BasedModel ............................................................. 107 OmarJaaforandBabigaBirregah ThroughtheGrapevine:AComparisonofNewsinMicroblogs andTraditionalMedia........................................................... 135 ByungkyuKang,HaleighWright,TobiasHöllerer,AmbujK.Singh, andJohnO’Donovan Prediction of Elevated Activity in Online Social Media Using AggregatedandIndividualizedModels........................................ 169 JimpeiHarada,DavidDarmon,MichelleGirvan,andWilliamRand Unsupervised Link Prediction Based on Time Frames inWeighted–DirectedCitationNetworks..................................... 189 MehmetKaya,MujtabaJawed,ErtanBütün,andRedaAlhajj ix x Contents AnApproachtoMaximizetheInfluenceSpreadintheSocialNetworks.. 207 IbrahimaGaye,GervaisMendy,SamuelOuya,andDjarafSeck EnergyEfficiencyAnalysisoftheVeryFastDecisionTreeAlgorithm..... 229 EvaGarcia-Martin,NiklasLavesson,andHåkanGrahn Glossary........................................................................... 253 Contributors Reda Alhajj Department of Computer Science, University of Calgary, Calgary, AB,Canada KorraSathyaBabu NationalInstituteofTechnologyRourkela,Rourkela,India NadineBaptiste-Jessel ToulouseInstituteofComputerScienceResearch(IRIT), UniversityofToulouse,CNRS,INPT,UPS,UT1,UT2J,Toulouse,France Santosh Kumar Bharti National Institute of Technology Rourkela, Rourkela, India Babiga Birregah Charles Delaunay Institute, UMR CNRS 6281, University of TechnologyofTroyes,Troyes,France Ertan Bütün Department of Computer Engineering, Fırat University, Elazıg˘, Turkey Marie-FrançoiseCanut ToulouseInstituteofComputerScienceResearch(IRIT), UniversityofToulouse,CNRS,INPT,UPS,UT1,UT2J,Toulouse,France David Darmon Department of Mathematics, University of Maryland, College Park,MD,USA DavidN.Fisher DepartmentofIntegrativeBiology,UniversityofGuelph,Guelph, ON,Canada DanielW.Franks DepartmentofBiology,UniversityofYork,York,UK DepartmentofComputerScience,UniversityofYork,York,UK EvaGarcia-Martin BlekingeInstituteofTechnology,Karlskrona,Sweden IbrahimaGaye UCAD-ESP,LIRTofSénégal,Dakar,Fann,Senegal Michelle Girvan Department of Physics, University of Maryland, College Park, MD,USA HåkanGrahn BlekingeInstituteofTechnology,Karlskrona,Sweden xi
The list of books you might like
