Francesco Ricci Lior Rokach Bracha Shapira   Editors Recommender Systems Handbook Third Edition Recommender Systems Handbook Francesco Ricci • Lior Rokach (cid:129) Bracha Shapira Editors Recommender Systems Handbook Third Edition Editors FrancescoRicci LiorRokach FacultyofComputerScience SoftwareandInformationSystems FreeUniversityofBozen-Bolzano Engineering Bozen-Bolzano,Italy Ben-GurionUniversityoftheNegev Beer-Sheva,Israel BrachaShapira SoftwareandInformationSystems Engineering Ben-GurionUniversityoftheNegev Beer-Sheva,Israel ISBN978-1-0716-2196-7 ISBN978-1-0716-2197-4 (eBook) https://doi.org/10.1007/978-1-0716-2197-4 ©SpringerScience+BusinessMedia,LLC,partofSpringerNature2011,2015,2022 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,expressedorimplied,withrespecttothematerialcontainedhereinorforany errorsoromissionsthatmayhavebeenmade.Thepublisherremainsneutralwithregardtojurisdictional claimsinpublishedmapsandinstitutionalaffiliations. ThisSpringerimprintispublishedbytheregisteredcompanySpringerScience+BusinessMedia,LLC, partofSpringerNature. Theregisteredcompanyaddressis:1NewYorkPlaza,NewYork,NY10004,U.S.A. Dedicatedto ourfamiliesinappreciationfortheirpatienceand supportduringthepreparationofthishandbook andto allourstudentsinappreciationoftheirideas, patience,andstimulusforbetterunderstanding thetopicscoveredinthishandbook F.R. L.R. B.S. Preface Recommendersystemsaresoftwaretoolsandtechniquesprovidingsuggestionsfor itemstobeofusetoauser.Thesuggestionsprovidedbyarecommendersystemare aimedatsupportingtheirusersinvariousdecision-makingprocesses,suchaswhat itemstobuy,whatmusictolisten,orwhatnewstoread.Recommendersystemsare valuable means for online users to cope with information overload and help them makebetterchoices.Theyarenowoneofthemostpopularapplicationsofartificial intelligence, supporting information discovery on the Web. Several techniques for recommendation generation have been proposed, and during the last two decades, many of them have also been successfully deployed in commercial environments. Nowadays,allthemajorInternetplayersadoptrecommendationtechniques. Development of recommender systems is a multi-disciplinary effort which involvesexpertsfromvariousfieldssuchasartificialintelligence,humancomputer interaction, data mining, statistics, decision support systems, marketing, and con- sumerbehavior. The first two editions of the handbook, which were published 10 and 6 years ago, were extremely well received by the recommender systems community. The positive reception, along with the fast pace of research in recommender systems, motivated us to further update the handbook. This third edition aims at updating the previously presented material and to show new techniques and applications in thefield.TheRecommenderSystemsHandbookisnowofferedinagreatlyrevised edition;11chaptersaretotallynew,andtheremainingchaptersareupdatedversions ofselectedchaptersalreadypublishedinthesecondedition. Despite these revisions, the goal of this handbook remains unaltered. It still aims at presenting both fundamental knowledge and more advanced topics by organizingtheminacoherentandunifiedrepositoryofrecommendersystemsmajor concepts, theories, methods, trends, challenges, and applications. Its informative, factual pages will provide researchers, students, and practitioners in industry with a comprehensive, yet concise and convenient reference source to recommender systems. Thebookdescribesindetailtheclassicalmethodsaswellasextensionsandnovel approaches that were more recently introduced. It consists of five parts: General vii viii Preface Recommendation Techniques; Special Recommendation Techniques; Value and ImpactofRecommenderSystems;HumanComputerInteraction;andApplications. Thefirstpartpresentsthemostpopularandfundamentaltechniquesusednowadays for building recommender systems, such as collaborative filtering, semantic-based methods, recommender systems based on implicit feedback, neural networks, and context-awaremethods.Thesecondpartcompriseschaptersonmoreadvancedrec- ommendationtechniques,suchassession-basedrecommendersystems,adversarial machine learning for recommender systems, group recommendation techniques, reciprocal recommenders systems, natural language techniques for recommender systems, and cross-domain approaches to recommender systems. The third part coversawideperspectiveontheevaluationofrecommendersystemswithchapters on methods for evaluating recommender systems, their value and impact, the multistakeholder perspective of recommender systems, and the analysis of the fairness, novelty, and diversity in recommender systems. The fourth part contains a few chapters on the human computer dimension of recommender systems, with papersontheroleofexplanation,theuserpersonality,andhowtoeffectivelysupport individualandgroupdecisionwithrecommendersystems.Thelastpartfocusseson applicationinseveralimportantareas,suchas,food,music,fashion,andmultimedia recommendation. We would like to thank all authors for their valuable contributions. We would like to express gratitude to all the reviewers who generously provided comments on drafts or counsel otherwise. We would like to express our special thanks to Susan Lagerstrom-Fife and staff members of Springer for their kind cooperation throughout the production of this book. Finally, we wish this handbook will contribute to the growth of this subject; we wish the novices a fruitful learning path, and to those more expert, a compelling application of the ideas discussed in thishandbookandafruitfuldevelopmentofthischallengingresearcharea. Bozen-Bolzano,Italy FrancescoRicci Beer-Sheva,Israel LiorRokach Beer-Sheva,Israel BrachaShapira February2022 Contents RecommenderSystems:Techniques,Applications,andChallenges ..... 1 FrancescoRicci,LiorRokach,andBrachaShapira PartI GeneralRecommendationTechniques Trust Your Neighbors: A Comprehensive Survey of Neighborhood-BasedMethodsforRecommenderSystems ............... 39 Athanasios N. Nikolakopoulos, Xia Ning, Christian Desrosiers, andGeorgeKarypis AdvancesinCollaborativeFiltering ......................................... 91 YehudaKoren,SteffenRendle,andRobertBell ItemRecommendationfromImplicitFeedback............................ 143 SteffenRendle DeepLearningforRecommenderSystems.................................. 173 ShuaiZhang,YiTay,LinaYao,AixinSun,andCeZhang Context-Aware Recommender Systems: From Foundations toRecentDevelopments....................................................... 211 Gediminas Adomavicius, Konstantin Bauman, Alexander Tuzhilin, andMosheUnger SemanticsandContent-BasedRecommendations.......................... 251 CataldoMusto,MarcodeGemmis,PasqualeLops,FedelucioNarducci, andGiovanniSemeraro PartII SpecialRecommendationTechniques Session-BasedRecommenderSystems....................................... 301 DietmarJannach,MassimoQuadrana,andPaoloCremonesi ix x Contents AdversarialRecommenderSystems:Attack,Defense,andAdvances ... 335 Vito Walter Anelli, Yashar Deldjoo, Tommaso Di Noia, andFeliceAntonioMerra GroupRecommenderSystems:BeyondPreferenceAggregation ........ 381 JudithMasthoffandAmraDelic´ People-to-PeopleReciprocalRecommenders ............................... 421 IrenaKoprinskaandKalinaYacef NaturalLanguageProcessingforRecommenderSystems ................ 447 OrenSarShalom,HaggaiRoitman,andPigiKouki DesignandEvaluationofCross-DomainRecommenderSystems........ 485 MaurizioFerrariDacrema,IvánCantador,IgnacioFernández-Tobías, ShlomoBerkovsky,andPaoloCremonesi PartIII ValueandImpactofRecommenderSystems ValueandImpactofRecommenderSystems ............................... 519 DietmarJannachandMarkusZanker EvaluatingRecommenderSystems .......................................... 547 AselaGunawardana,GuyShani,andSivanYogev NoveltyandDiversityinRecommenderSystems........................... 603 PabloCastells,NeilHurley,andSaúlVargas MultistakeholderRecommenderSystems................................... 647 HimanAbdollahpouriandRobinBurke FairnessinRecommenderSystems .......................................... 679 MichaelD.Ekstrand,AnubrataDas,RobinBurke,andFernandoDiaz PartIV HumanComputerInteraction BeyondExplainingSingleItemRecommendations ........................ 711 NavaTintarevandJudithMasthoff PersonalityandRecommenderSystems..................................... 757 MarkoTkalcˇicˇ andLiChen IndividualandGroupDecisionMakingandRecommenderSystems.... 789 AnthonyJameson,MartijnC.Willemsen,andAlexanderFelfernig PartV RecommenderSystemsApplications SocialRecommenderSystems ................................................ 835 IdoGuy FoodRecommenderSystems ................................................. 871 DavidElsweiler,HannaHauptmann,andChristophTrattner Contents xi MusicRecommendationSystems:Techniques,UseCases,and Challenges....................................................................... 927 MarkusSchedl,PeterKnees,BrianMcFee,andDmitryBogdanov MultimediaRecommenderSystems:AlgorithmsandChallenges........ 973 Yashar Deldjoo, Markus Schedl, Balázs Hidasi, Yinwei Wei, andXiangnanHe FashionRecommenderSystems.............................................. 1015 Shatha Jaradat, Nima Dokoohaki, Humberto Jesús Corona Pampín, andRezaShirvany Index............................................................................. 1057