Quantifying and suppressing ranking bias in a large citation
network
Giacomo Vaccario
a, Matúˇs Medo
b,c,d, Nicolas Wider
a,
Manuel Sebastian Mariani
d,e,∗aChair of Systems Design, ETH Zurich, 8092 Zurich, Switzerland
bInstitute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu 610054, PR China cDepartment of Radiation Oncology, Inselspital, Bern University Hospital and University of Bern, 3010 Bern, Switzerland
dDepartment of Physics, University of Fribourg, 1700 Fribourg, Switzerland
eGuangdong Province Key Laboratory of Popular High Performance Computers, College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, PR China
Keywords: Impact indicators Ranking Network analysis Field bias Field normalization
It is widely recognized that citation counts for papers from different fields cannot be directly compared because different scientific fields adopt different citation practices. Cita-tion counts are also strongly biased by paper age since older papers had more time to attract citations. Various procedures aim at suppressing these biases and give rise to new normalized indicators, such as the relative citation count. We use a large citation dataset from Microsoft Academic Graph and a new statistical framework based on the Mahalanobis distance to show that the rankings by well known indicators, including the relative citation count and Google’s PageRank score, are significantly biased by paper field and age. Our statistical framework to assess ranking bias allows us to exactly quantify the contributions of each individual field to the overall bias of a given ranking. We propose a general nor-malization procedure motivated by the z-score which produces much less biased rankings when applied to citation count and PageRank score.
1. Introduction
Paper citation count itself and various quantities derived from it are used as influential indicators of research impact (Garfield, 2006; Hirsch, 2005). At the same time, it is well known that the cumulative number of citations received by academic publications strongly depends on paper age and field (Schubert & Braun, 1986; Vinkler, 1986). Old papers have had more time to acquire citations than recent ones, and their advantage is further enhanced by the preferential attachment mechanism (de Solla Price, 1976; Newman, 2009). While heterogeneous paper fitness and paper aging possibly attenuate the advantage of old nodes (Medo, Cimini, & Gualdi, 2011; Wang, Song, & Barabási, 2013), empirical evidence typically shows that citation count is still biased toward old nodes (seeMariani, Medo, & Zhang, 2016;Newman, 2009; Radicchi & Castellano, 2011, among others). In addition, different academic fields adopt very different citation practices (seeBornmann & Daniel, 2008for a review on the topic), which results in a strong dependence of the mean number of citations on academic field, as shown in several works (Bornmann & Daniel, 2009; Lundberg, 2007;Radicchi, Fortunato, & Castellano, 2008, among others).
∗ Corresponding author at: Department of Physics, University of Fribourg, 1700 Fribourg, Switzerland. E-mail address:manuel.mariani@unifr.ch(M.S. Mariani).
http://doc.rero.ch
Published in "Journal of Informetrics 11(3): 766–782, 2017"
which should be cited to refer to this work.
Anaturalquestionarises:howcanwe“fairly”usecitation-basedindicatorstocomparepapersfromdifferentfieldsandof differentage?Theproblemofcomparingpapersfromdifferentfieldsisusuallyreferredtoasthefield-normalizationproblem. Severalapproachestoaddressthisquestionhavebeenproposedintheliterature(seeWaltman,2016forarecentreview).A particularlysimpleapproachistodivideeachpaper’scitationcountbythemeannumberofcitationsforpapersofthesame fieldpublishedinthesameyear.TheresultsbyRadicchietal.(2008)suggestedthatthisindicator,calledrelativecitation count,producesarankingthatisstatisticallyconsistentwiththehypothesisofarankingthatisnotbiasedbyfieldandage. ThisfindinghasbeenchallengedbysubsequentworksbyAlbarrán,Crespo,Ortuno,andRuiz-Castillo(2011)andWaltman, vanEck,andvanRaan(2012),whichleavesthedebateonage-andfield-normalizationproceduresstillopen.
In thisarticle,weanalyzealargedatasetfromMicrosoftAcademicGraph(Sinhaetal.,2015)toshowthatexisting indicatorsofimpact,includingtherelativecitationcount,failtoproducerankingsthatarenotbiasedbyageandfield.To simultaneouslyassessthesebiases,wepresentanewprocedurebasedontheMahalanobisdistance(Mahalanobis,1936). Thispermitstocomparetherankingbyagivenindicatorwiththoseobtainedwithasimulatedunbiasedsampling,andhence toquantifytheoverallrankingbias.Ananalyticresultderivedinthispaperallowsustoassessthecontributionofeachfield totheoverallrankingbias.Itisworthnoticingthatwhilewefocusonthebiasesbyageandfield,ourbiasassessment procedurecanbeeasilyextendedtodetectanyotherkindofinformationbias.
WealsopresentthefirstsystematicstudyofthepossiblebiasbyfieldofthePageRankscore(Brin&Page,1998)and ofitsage-rescaledversionintroducedbyMarianietal.(2016)atarticle-level.Themotivationtoanalyzethese network-basedindicatorscomesfromthefindingthattheyoutperformothermetricsinidentifyingexpert-selectedmilestonepapers (Marianietal.,2016).However,theapplicationofPageRankanditsvariantstoacademiccitationnetworksfocusedon datasetscomposedofpapersfromasinglefield(Chen,Xie,Maslov,&Redner,2007;Marianietal.,2016;Mariani,Medo, &Zhang,2015;Walker,Xie,Yan,&Maslov,2007;Yao,Wei,Zeng,Fan,&Di,2014;Zhou,Zeng,Fan,&Di,2016).While thepossiblebiasbyscientificfieldofeigenvector-basedalgorithmshasbeenexploredbyWaltmanandvanEck(2010)at journal-level,thePageRankscore’spossiblebiasbyacademicfieldatarticle-levelis(toourbestknowledge)stillunexplored andwearethefirstauthorstoaddressit.
Weintroducetwonovelindicatorsofimpactmotivatedbythez-score:age-andfield-rescaledcitationcountRAF(c)and
age-andfield-rescaledPageRankRAF(p).Wefindthatthenovelindicatorsproducepaperrankingsthataremuchlessbiased
byageandfieldthantherankingsproducedbytheotheranalyzedindicators.Nevertheless,alsotheMahalanobisdistance observedforthenewindicatorsisnotstatisticallyconsistentwithonesobtainedforasimulatedunbiasedprocess.This indicatesthattheproblemofachievinganidealunbiasedrankingofthepublicationsremainsopen.
Therestofourarticleisorganizedasfollows:Section2describestheanalyzeddatasetofpublicationsobtainedfrom theMicrosoftAcademicGraph.Section3presentsexistingpaper-levelimpactindicatorsandreportstheirbiasbyscientific field.InSection4,weintroducearescalingprocedureforcitationcountandPageRankscoresmotivatedbythez-score.In Section5,weintroduceageneralproceduretotestforanykindofrankingbias,andpresentitsapplicationtoassessthefield andagebiasoftherankingsbytheindicatorsstudiedhere.InSection6,weconcludebydiscussingpossiblelimitationsof ouranalysisandfutureresearchdirections.
2. Data
WeanalyzeabibliographicdatasetwhichwasprovidedfortheKDDCup2016.1 ThisdataisadumpoftheMicrosoft
AcademicGraph(MAG)andcontainsmorethan126millionsofpublicationsandmorethan467millionscitations(Sinha etal.,2015).EachpublicationisalsoendowedwithvariouspropertiessuchasuniqueID,publicationdate,titleandjournal ID.Wepre-processedthedata(detailsareprovidedinAppendixA)toremovefromtheanalysispaperswithincomplete information,endingupwithN=18193082uniquepublicationsandE=109719182citations.
TheMAGhasafieldclassificationatpaperlevel(Sinhaetal.,2015).IntheKDDcupdump,thereare19mainfieldsand numeroussubfieldsupto3hierarchicallevelsofsubsubfields.However,allthedifferentsubfieldscanbelongtoseveral mainfields,meaningthateachpublicationcanbelongtomorethanonemainfield.Weuseherethefieldclassificationat thehighesthierarchicallevel,i.e.,weonlyconsiderthe19mainfields.WhencalculatingthecitationcountandPageRank scoreofpapers(seeSection3),weconsiderthepublicationsthatbelongtomorethanonefieldonlyonce.Inthisway,wedo notmodifythenumberofcitationsthateachpaperreceivesandgives,andwedonotchangethetopologyofthenetwork onwhichthePageRankscoresarecalculated.Ontheotherhand,inagreementwithWaltmanetal.(2012),tocompute thefields’size(seeTableA.2inAppendixA)andthefield-rescaledmetrics(seeSections3and4),eachpublicationcanbe consideredmultipletimesintheanalysis,onceforeachfieldthepublicationbelongsto.Inthisway,eachfieldisrepresented byallitspublicationsevenifsomeofthesearesharedwithotherfields.
BeforemovingtothenextSections,wedevoteourattentiontotwomainassumptionsofouranalysis.First,weassume thattheMicrosoftAcademicdataprovidearepresentativesampleofthepopulationofpublicationsandoftheircitations.This assumptionismotivatedbythefindingsofindependentanalysesoftheMicrosoftAcademicdataset(Harzing&Alakangas, 2013;Hug&Braendle,2017)thathaveshownthatitscoverageiscomparabletootherpopularacademicdatabases,suchas ScopusandWebofScience.
1https://kddcup2016.azurewebsites.net/Data
Second,toquantifythebiasbyfieldofimpactindicators,weassumethatthefieldsaregivenbytheMicrosoftAcademic’s fieldclassificationschemeatitshighesthierarchicallevel.Intheliterature,thereisnogeneralagreementonwhichfield classificationschemeshouldbeusedtoclassifypapersandthereisanentirestreamofworksinvestigatingissuesrelatedto this(Adams,Gurney,&Jackson,2008;Colliander&Ahlgren,2011;Radicchi&Castellano,2012a;Sirtes,2012;Zitt, Ramanana-Rahary,&Bassecoulard,2005).In particular,thechoiceofa suitableaggregationlevelhasbeenshown tobedelicate: byconsideringthemostaggregatefields,heterogeneitiesinthesubfields’citationpatternsmightbehidden(Radicchi& Castellano,2011;vanLeeuwen&Medina,2012)–thiseffecthasbeenshowntobemagnifiedwheniterativeranking algorithmsareusedinsteadofcitationcount(Waltman,Yan,&vanEck,2011).Ontheotherhand,increasingtheresolution ofthefieldclassificationmayleadtolargely-overlappingfieldsortohardlyinterpretablefields.Forexample,Hug,Ochsner, andBrandle(2017)showthattheMAGfieldsatthesecondhighestlevelaretoodetailedand,forthisreason,theauthors suggestthattheyshouldnotbeusedforfield-normalizationpurposes.Weleavetofutureresearchtheimportantstudyof howdifferentclassificationschemesimpactthebiasesofrankingsandhowourresultsgeneralizetootherdatasets.
3. Shortcomingsofexistingmetrics
Wenowdefinethefourexistingmetricsanalyzedinthiswork:citationcountc,relativecitationcountcf,PageRankscore
p,age-rescaledPageRankscoreRA(p)(Section3.1).Furthermore,weshowthatthesemetricsareseverelybiasedbyscientific
field(Section3.2).
3.1. Definitionofexistingmetrics
Citationcount,c.Thecitationcountciofnodeiissimplythenumberofcitationsreceivedbypaperi.Intermsofthecitation
network’sadjacencymatrixA(inadirectednetwork,Aij=1ifnodejpointstonodei,Aij=0otherwise),wecanexpressthe
citationcountasci=
jAij.
Relativecitationcount,cf.Toovercomecitationcount’sbiasbypaperageandacademicfield,Radicchietal.(2008)defined
therelativecitationcountcifofpaperiascif:=ci/Yi(c),whereYi(c)denotesthemeancitationcountforpaperspublished
inthesamefieldandyearaspaperi.Throughoutthispaper,wealwaysrefertothe19mainfieldsprovidedintheMAG dataset.
PageRank,p.Citationcountandmetricsbuiltonitshareanimportantlimitation:thecitationsapaperreceivesareall countedthesame,regardlessoftheimportanceofthecitingpaper.Apossiblewaytoovercomethislimitation–recognized alreadyinthe70sinthescientometricscommunity(Pinski&Narin,1976)–istotakeintoaccountthewholestructureof thepaper-papercitationnetwork.Inthisspirit,eigenvector-basedmetricstakeasinputthecitationnetwork’sadjacency matrixA.Thisclassofmetricshavebeenappliedinvariousresearchdomainsincludingscientometrics(Bergstrom,West, &Wiseman,2008;Pinski&Narin,1976),Webinformationretrieval(Brin&Page,1998;Kleinberg,1999),socialscience (Bonacich,1987;Katz,1953)–see(Ermann,Frahm,&Shepelyansky,2015;Franceschet,2011;Gleich,2015)forareview. Amongthesemetrics,wefocusonGoogle’sPageRankscore(Brin&Page,1998).Thiswasoriginallydevisedtorankwebpages intheWorldWideWebandhasattractedconsiderableinterestofthescientometricscommunity.Therationalebehindits applicationtocitationnetworksisthatcitationscomingfrominfluentialpapersshouldcountmorethancitationsfrom obscurearticles.
ThePageRankscoresofpapersarecontainedinavectorpdefinedbythefollowingequation
p=˛Pp+(1−˛)v, (1)
where˛isaparameterofthealgorithm(calleddampingfactor),Pistherandom-walktransitionmatrixwithelements Pij=Aij/koutj ,koutj =
lAlj isthenumberofreferencesinpaperj,andvisauniformteleportationvectorwithelements
v
i=1/Nforallpapersi.Eq.(1)canbeinterpretedasthestationaryequationofastochasticprocessonthecitationnetwork. Inthisprocess,arandomwalkerisplacedoneachpaperandhe/sheeitherfollowsacitationedgewithprobability˛,or jumpstoarandomlychosenpaperwithprobability1−˛.Whenthenumberofwalkersoneachpaperreachesastationary value,weobtainthePageRankscoreofapaperibycalculatingthefractionofwalkeronthispaper.Thereisnouniversal criteriontochoosethevalueofthedampingfactor˛.InagreementwithChenetal.(2007),weset˛=0.5whichcorresponds toarandomwalkercoveringpathsoflengthtwobeforeteleportingtoarandomnode,asopposedtopathsoflengthcloseto sevenexpectedwiththeoftenusedvalue˛=0.85.Chenetal.(2007)arguethatthechoice˛=0.5betterreflectstheactual surfingbehaviorofresearchersthan˛=0.85.PageRankisbasedonastatic,time-aggregatedperspectiveoftheconsiderednetwork.Ingeneral,suchperspectivehas beenshowntobelimitingfortheanalysisofevolvingnetworks(Marianietal.,2015;Scholtes,Wider,Pfitzner,Garas,Tessone, &Schweitzer,2014b,Wider,etal.,2014).Whiletheresultingmetric’sbiastowardsoldpapershasalreadybeenstudiedin theliterature(Chenetal.,2007;Marianietal.,2015,2016;Maslov&Redner,2008),itspossiblebiasbyacademicfieldisstill unexploredandweaddressitinSection3.2.
Age-rescaledPageRank,RA(p).TosuppresstheagebiasofPageRank,Marianietal.(2016)proposedtorescalethePageRank
scorebycomparingeachpaper’sscorewiththescoresofpapersofsimilarage.Assumingthatthepapersareorderedby
Fig.1.Fieldbiasoftheanalyzedcitation-basedmetrics.Toppanelsshowhistogramsofthefractionoftop-1%publicationsforeachfieldintherankingby (lefttoright)citationcountandrelativecitationcount.Theblackhorizontallineisat0.01,i.e.theexpectedvalue.Bottompanelsshowforeachfieldthe complementarycumulativedistributionsforcitationcount(left)andrelativecitationcount(right).
oldertoyounger,onecomputesthemeanvalueA
i(p)andthestandarddeviationiA(p)ofPageRankscoresoverppapers
aroundi,i.e.j∈[i−p/2,i+p/2].Consequently,therescaledPageRankscoreRAi(p)ofpaperiisdefinedas
RA i(p)= pi−Ai(p) A i(p) . (2)
Marianietal.(2016)appliedrescaledPageRank tothenetwork ofphysicspapers publishedbytheAmericanPhysical Societyjournalstoshowthattheresultingrankingisnotbiasedbypaperageand,asaresult,itallowsustoidentifyseminal publicationmuchearlierthanrankingsbymetricsthatarebiasedagainstrecentpapers.Inthefollowing,wesetp=1000
asinMarianietal.(2016).
3.2. Fieldbiasoftheexistingmetrics
Afterhavingdescribedasetofexistingmetrics,wenowapplythemtotheMAGdatasettoshowthattherankingsthat theyproducearebiasedbyscientificfield.Forarankingthatisnotbiasedbyscientificfield,thenumberoftop-ranked publicationsfromeachfieldshouldbeproportionaltothetotalnumberofpublicationsfromthatfield.Inotherwords,for anunbiasedranking,weexpect
f =100z Kf (3)
papersfromfieldfamongthetopz%papersintheranking,whereKfisthetotalnumberofpublicationsfromfieldf(Radicchi
etal.,2008).Inthefollowing,wedenotebyk(m)f thenumberofpublicationsfromfieldfinthetop-1%oftherankingbymetric m.Werestrictouranalysistoz%=1%;resultsforothervaluesofzareavailableuponrequestfromtheauthors.
InthetoppanelsofFig.1,weillustratethefieldbiasofcitationcount,c,andrelativecitationcount,cf.Thepresenceof
strongbiasesisevidentforbothmetricsbecausetherearefieldswhoseratiok(m)f /Kf isfarawayfromtheexpectedvalue 0.01.Inparticular,EnvironmentalScienceisextremelyover-representedinthetopoftherankingbycitationcount.We arguethatthisbiascomesfromthefactthatpublicationsfromthisfieldhaveameancitationcountalmosttwiceasbig comparedtopublicationsbelongingtootherfields(seeTableA.2).Forrelativecitationcount,wefindabetteragreement withwhatwewouldexpectfromanunbiasedindicator.However,relativelylargedeviationsarestillevident,especiallyfor thefieldofPoliticalScience.
InthebottompanelsofFig.1,wereportthedistributionsofcandcfforeachfield.Thesepanelsshowthatthebias
byfieldisnotlimitedtothetop1%papersintheranking,butitarisesfromsystematicdifferencesbetweenthescore distributionsacrossdifferentfields.Forexample,whenlookingatthedistributionofc,papersinthefieldofPoliticalScience haveconsistentlysmallerprobabilitytohavemorethanonecitationcomparedtootherfields.Foradetaileddiscussion aboutthebiasoftherankingbycf,werefertoAppendixC.
Fig.2reportsthesameanalysisforPageRankscores,p,andage-rescaledPageRankscores,RA(p).Thisfigureprovidesthe
firststudyofthedependenceofPageRankscoreonacademicfield.ThetoppanelsofFig.2showthatthetoppositionsofboth rankingsarebiasedbyfield,andbothrankingsoverestimatetheimpactofpublicationsinthefieldofEnvironmentalScience. Again,wearguethatthishappensbecausethemeanindegreeofpublicationsfromEnvironmentalScienceisapproximately twiceasbigcomparedtopublicationsthatbelongtootherfields(seeTableA.2).Fromthebottom-leftpanelofFig.2,wefind thatthefulldistributionofscoresofPageRankhaveasimilarshape,butdifferentbroadness.Thesedifferencesareslightly
Fig.2. FieldbiasoftheanalyzedmeasuresbasedonPageRank.Toppanelsshowhistogramsofthefractionoftop-1%publicationsforeachfieldinthe rankingby(lefttoright):PageRankandage-rescaledPageRank.Theblackhorizontallineisat0.01,i.e.theexpectedvalue.Bottompanelsshowforeach fieldthecomplementarycumulativedistributionsforPageRank(left)andage-rescaledPageRank(right).
smallerfortheage-rescaledPageRank,withtheexceptionofthefieldofEnvironmentalScience(seebottomrightpanelof Fig.2).
4. Definingnewage-andfield-normalizedmetrics
Inthissection,weintroducetwonovelindicatorsofpaperimpact:theage-andfield-rescaledcitationcount,RAF(c),
andtheage-andfield-rescaledPageRank,RAF(p).Thetwoindicators,RAF(c)andRAF(p),areobtainedfromcitationcount
candPageRankscorep,respectively,througharescalingprocedure.Thisprocedureisbasedonthez-scoreandisaimed atsuppressingageandfieldbias.Theideaofusingthez-scoreisnotnewinscientometrics(Bornmann&Daniel,2009; Lundberg,2007;Marianietal.,2016;McAllister,Narin,&Corrigan,1983;Newman,2009;Zhang,Cheng,&Liu,2014);our newindicatorscanbeconsideredasvariantsoftheindicatorbasedonthez-scorestudiedbyZhangetal.(2014)andtheir maindifferenceisexplainedbelow.
4.1. Age-andfield-rescaledcitationcount,RAF(c)
Tocalculatetheage-andfield-rescaledcitationcountRAF
i (c)ofapaperibelongingtoafieldf,wefirstcomputethemean
AF
i (c)andthestandarddeviationiAF(c)ofthecitationcountofpapersofthesamefieldandofsimilarageaspaperi.In
particular,AF
i (c)andAFi (c)arecomputedoverthepapersthatbelongtothesamefieldfaspaperiandthatareamongthe
cclosestpaperstoiasmeasuredbythedistance|i−j|betweentheirrankbyage.Then,theage-andfield-rescaledcitation
countscoreRAF i (c)isdefinedas RAF i (c)= ci−AFi (c) AF i (c) . (4)
Theaveragingwindowsizecisaparameterofthemethod,whichwesettoc=1000.
DifferentlyfromZhangetal.(2014),forthecomputationofthez-score,weusetemporalwindowswiththesamenumber ofpublications,whichingeneralcorrespondstoreal-timeintervalsofdifferentduration.Thischoiceissupportedbyrecent findings(Paroloetal.,2015)thatindicatethatincitationnetworks,timeisbetterdefinedbynumberofpublicationsthan byrealtime.Furthermore,rescaledmetricsbasedonthez-scorewithfixedtemporal-windowdurationhavealreadybeen showntounderperformwithrespecttotherelativecitationcountcfinthetaskofproducinganunbiasedranking(Zhang
etal.,2014).Ouranalysis(notshown)confirmsthatwhenthetemporalwindowsareofafixedtemporallength,thebias removalisinferiortothatachievedwithtemporalwindowscontainingafixednumberofpublications.Forthesereasons, wedonotincludemetricsbasedonz-scorewithfixedtemporal-windowdurationinouranalysis.
Differentlyfromtherelativecitationcountcf,RAF(c)isexpectedtohavenotonlyuniformmeanvalueacrossdifferent
publicationdatesandfields,butalsouniformstandarddeviation.Thisshouldleadtoamorebalancedrankingofthepapers. Weshowinthefollowingthatthisisindeedthecase.
Fig.3. Fieldbalanceoftheanalyzedcitation-countandPageRank-basedmetrics.Toppanelsshowhistogramsofthenumberoftop-1%publicationsfor eachfieldintherankingby(lefttoright):age-andfield-rescaledcitationcount,andage-andfield-rescaledPageRank.Theblackhorizontallineisat0.01, i.e.theexpectedvalue.Bottompanelsshowforeachfieldthecomplementarycumulativedistributionsforage-andfield-rescaledcitationcount(left),and age-andfield-rescaledPageRank(right).
4.2. Age-andfield-rescaledPageRank,RAF(p)
PreviousworkshaveshownthatPageRankisbiasedtowardsoldpapersinscientificcitationnetworks(Chenetal.,2007; Marianietal.,2015;Maslov&Redner,2008).Moreover,wehaveshowninSection3thatPageRankscorepisbiasedby scientificdomain.Tosimultaneouslysuppressthesetwobiases,weproposetheage-andfield-rescaledPageRankscore RAF(p).RAF(p)isdefinedsimilarlyasRAF(c):wecomputethemeanvalueAF
i (p)andthestandarddeviationiAF(p)ofthe
PageRankscoresofthepapersthatbelongtothesamefieldaspaperiandthatareamongthepclosestpaperstoias
measuredbythedistance|i−j|betweentheirrankbyage.Theage-andfield-rescaledPageRankscoreisthendefinedas RAFi (p)=pi−AFi (p)
AF i (p)
. (5)
Inthefollowing,wesetp=1000.
4.3. Fieldbiasofthenewmetrics
In thetoppanelsofFig.3,weshowthatinthetop-1%oftherankingsbyRAF(p)andRAF(c)eachfield appearswell
represented.Infact,thedeviationsfromtheexpectedvalueareverysmallespeciallyifcomparedtothedeviationsofthe otherrankings(seetoppanelsinFigs.1and2).InthebottompanelsofFig.3,wereportthatthefullscoredistributionsfor papersfromdifferentfieldscollapseextremelywelltopofeachotherthankstotherescalingprocedure.
5. Quantifyingrankings’biasesbyfieldandage
WebeginthisSectionbyintroducinganewmethodologytoassessaranking’sbiasbasedontheMahalanobisdistance (Section5.1).Then,weusethistoquantifythebiasbyfield(Section5.2)andthebiasbyageandfield(Section5.3). 5.1. AgeneralframeworktoassessrankingbiasesbasedontheMahalanobisdistance
WhileFigs.1and2illustratethesubstantialfieldbiasoftheexistingmetrics,thebiasismuchweaker(ifany)forthe newage-andfield-rescaledmetricsinFigure3.Nowwequantifythisimprovementbyextendingthestatisticaltestsofbias suppressionpresentedbyMarianietal.(2016),RadicchiandCastellano(2012b),Radicchietal.(2008),Waltmanetal.(2012). Similarlytotheseworks,weassumethatarankingisunbiasedifitspropertiesareconsistentwiththoseofanunbiased selectionprocess.
5.1.1. Assessingthebiasbyfield
Letusfirstanalyzetheproblemofassessingthebiasbyfield.ConsideranurnwhichcontainsNmarbles,eachofthem correspondingtooneofthepublicationspresentinourdataset.Anunbiasedselectionprocessthencorrespondstosampling fromthisurnatrandomwithoutreplacementafixednumbern=N×0.01ofpublications.Fromtheextractedsample,we countthenumberofpublicationsthatbelongtoeachfieldf,kf,andrecordthesenumbersinthevector k=(k1,...,kF)T;here
Fig.4.Mahalanobisdistances,dM,fortheanalyzedindicatorswhenconsideringthe19mainfields.Fromlefttoright:citationcount,relativecitationcount,
PageRank,age-rescaledPageRank,age-andfield-rescaledcitationcountandage-andfield-rescaledPageRank.Thehorizontalredlinerepresentstheupper boundofthe95%confidenceintervalobtainedfromthesimulations.Intheinsets,wereportthedistributionofdMcomingfrom1000000simulationsof
theunbiasedsamplingprocess.Again,theredlinerepresentstheupperboundofthe95%confidenceinterval.(Forinterpretationofthereferencestocolor inthisfigurelegend,thereaderisreferredtothewebversionofthearticle.)
Fdenotesthenumberoffields.Theprobabilitytoobserveacertainvector, k,isgivenbythemultivariatehypergeomentric distribution(MHD) P
k= F f Kf kf N n(6)
whereKfisthetotalnumberofpublicationsinfieldf.Followingthisselectionprocess,amongthenextractedpublications,
theexpectednumberofpublicationsforfieldfisf=nKf/N.
Assumethattheactualrankingbyagivenmetricmfeatureskf(m)publicationsfromfieldfinthetop1%ofitsranking.In general,theobservedk(m)f deviatesfromitsexpectedvaluef.Asimpleapproachtoquantifythisdeviationwouldconsist
incomputingthez-score,definedasz(m)f :=(k(m)f −f)/f,wherefistheexpectedstandarddeviationforfieldfaccording
totheMHDspecifiedbyEq.(6).Therearehowevertwoshortcomingsofthez-score.First,thez-scoreonlygivespartial informationforaMHD–howfarfromtheexpectedvaluesweareinunitsofstandarddeviations–butitdoesnotprovide informationonhowstatisticallysignificantthedeviationsare.Second,toquantifytheoverallbiasofagivenindicatorm, wewouldneedtoaggregatethez-scoresfromthedifferentfields.Forexample,wecouldtaketheaveragez-score,butthis wouldneglectthecorrelationbetweenthedifferentfieldscomingfromtheconstraintn=
fk(m)f .Toovercomethesetwoproblems,wefollowadifferentapproach.Wefirstrunvariousnumericalsimulationsthat repro-ducetheunbiasedselectionprocess.Thesesimulationsproduceasetofrankingvectorswhicharedistributedaccording toEq.(6)aroundthevectorofexpectedvalues, =(1,...,m).Differentlyfrom(Radicchi&Castellano,2012b),wedo
notestimatetheconfidenceintervalforthedifferentfieldsseparately.WecalculateinsteadtheMahalanobisdistance(dM, Mahalanobis(1936)andAppendixB)foreachsimulatedvectorfrom,andconstructthedistributionofdM’sobtainedby thesimulatedunbiasedselectionprocess.TheinsetoftheleftpanelofFig.4reportsthedistributionofthedMfor1000000 simulations.Thedistributioniscenteredarounditsmeanvalueof4.18andtheupperboundforthe95%confidencelevelis around5.37.2Foranidealunbiasedranking,wewouldexpectitsd
Mtofallintothe95%confidenceintervalofthedistribution ofthedMobtainedfromthesimulatedunbiasedsamplingprocess.
5.1.2. Assessingthebiasbyageandfield
Themethodologypresentedaboveiseasilygeneralizedtosimultaneouslyassessaranking’sbiasbyageandfield. Toaddthetemporaldimensiontothebiasassessmentprocedure,wesplitthepublicationsintoTequally-sizedagegroups, andrepeattheaboveanalysisbyusingF×Tdifferentcategoriesofpublications.InSection5.3,wesetF=19representing thenumberoffieldsandT=40asin(Marianietal.,2016),andthusweobtain760age-fieldgroupsofdifferentsizes.
2Acuriosityforthereader.Here,theaverageofthesquareofthed
Mfortheunbiasedsamplingprocessisextremelyclosetothenumberofdegreesof
freedomofourproblem.ThisstemsfromthefactthattheMHDdefinedbyEq.(6)convergestoaMultivariateGaussianDistribution(MGD)asweincrease thenumberofpublicationsNwhilekeepingn/Nfixedandsmall.Ourdatasetislargeenoughforthisapproximationtobeaccurate.Thed2
MofaMGDis
distributedasa2variablewithaverageequaltothenumberofdegreesoffreedom,i.e.18sincewehave19distinctfieldsandoneconstraint.
Table1
Theindividualcontributionz2
i(1−ki/N)ofeachfielditothedMofthedifferentmetrics.
Field c cf p RA(p) RAF(c) RAF(p) Art 1.15 1.95 3.01 0.31 2.84 0.09 Biology 36.46 15.81 6.74 0.87 0.12 0.16 Business 2.06 2.08 5.95 1.51 14.56 13.50 Chemistry 0.34 8.44 0.85 9.28 4.23 0.00 ComputerScience 0.29 23.86 0.02 3.09 0.12 13.50 Economics 3.34 6.51 4.20 5.77 32.32 7.93 Engineering 8.36 0.09 10.48 0.35 8.36 0.03 EnvironmentalScience 16.82 0.04 35.67 29.89 2.45 2.23 Geography 0.05 1.34 0.15 0.50 0.15 0.51 Geology 1.02 3.04 6.42 0.13 2.10 10.06 History 0.69 2.48 1.67 0.15 3.12 0.52 MaterialsScience 0.39 0.43 0.23 0.01 2.37 0.10 Mathematics 5.17 1.94 0.10 0.14 0.09 25.34 Medicine 4.02 7.66 0.06 1.94 2.16 19.56 Philosophy 1.49 3.23 0.12 0.36 0.15 0.07 Physics 8.30 0.21 6.00 33.67 14.34 0.01 PoliticalScience 1.47 10.22 6.82 0.51 1.47 2.44 Psychology 5.75 1.43 8.56 10.24 2.93 1.65 Sociology 2.83 9.23 2.95 1.30 6.11 2.30
TheboldvaluesmarkthefieldsthatgivethelargestcontributiontothedMofeachmetric.
5.2. Resultsonthebiasbyfield
TherankingsproducedbydifferentmetricsdiffergreatlybytheirdM(seeFig.4).Asexpected,themetricsthatarenot field-rescaled(c,RA(p),andp)arefarfrombeingunbiased.Atthesametime,relativecitationcountthatisrescaledbyfield
performsonlyslightlybetterthanPageRankwhichisignorantofanyfieldinformation.Thebestresultsbyawidemarginare achievedbyourmetrics,RAF(c)andRAF(p),obtainedusingthenewrescaling.Nevertheless,boththesemetricsfailtomeet
the95%upperboundachievedbysimulatedunbiasedrankings.Asdisappointingasitmayseem,thisfindingisnotentirely surprisingastheproposedrescalingproceduresfocusonequalizingthefirsttwomomentsoftherespectivequantities(c andp)whereasthequantities’distributionscandifferalsobyhighermoments.
TounderstandwhichfieldcontributesthemosttotheresultingdMvalues,wederivedanalternativeanalyticexpression forthedM dM(k, )2= F
i z2 i 1−ki N (7)whereweomitthemetricsuperscript (m)fromthenotationforziand k forsimplicity.Wehave proventhisformula
analyticallyforF=3,4,5,6,andwehavenumericallytesteditforF=19and760(seeAppendixB);itremainsopentoprove itinarbitrarydimensions.
InTable1,wereporttheindividualfields’contributionstothed2
McalculatedusingEq.(7).WefindthatBiologyand ComputerSciencearethefieldswhichgivethebiggestcontributionstothed2
Mfortherankingsbycitationcountand relativecitationcount,respectively.Thiscouldnothavebeendetectedbylookingatthedeviationsfromtheexpected values.Indeed,inFig.1weonlyseethatEnvironmentalScienceandPoliticalSciencehavethelargestdeviations.Forthe novelindicators,approximatelyonethirdofthed2
MofRAF(c)isexplainedbythefieldofEconomicsandapproximatelyone fourthofthed2
MofRAF(p)isexplainedbythefieldofMathematics.
Inaddition,wealsofindthatthedM’scontributionsacrossdifferentfieldsassumevaluesinarelativelybroadrange.This suggeststhatfindingsonrankings’biasbyfieldmaystronglydependonwhichdisciplinesareincludedornotintheanalysis. Wearguethatarbitrarychoicesonwhichfieldstoincludeshouldbeavoidedinfutureresearchonfield-normalizationof impactindicators.
Tosummarize,ourbiassuppressiontestallowsusnotonlytoestimatethelevelofbias(dM)ofthevariousmetrics,but alsotoquantifywhichpercentageofthetotalbias(d2
M)ofanindicatorisexplainedbyeachsinglefield. 5.3. Resultsonthebiasbyageandfield
Whiletheanalysisoftheprevioussectionfocusedontherankingbiasbyfield,inthissectionweusethedMto simulta-neouslyassessthebiasbyageandfieldofagivenranking.
InFig.5,weshowthedM’sforthedifferentindicatorsandforthe95%confidenceintervalforthesimulatedunbiased selectionprocessusing40×19age-fieldtypesofpublications.Forcitationcount,PageRank,relativecitationcountand age-rescaledPageRankwehavetorejectthehypothesisthattherankingsoftheseindicatorsarenotbiasedbyageandfield.For
Fig.5.Mahalanobisdistances,dM,fortheanalyzedindicatorswhenconsideringthe760age-fieldgroups.Fromlefttoright:citationcount,relativecitation
count,PageRank,age-rescaledPageRank,age-andfield-rescaledcitationcountandage-andfield-rescaledPageRank.Thehorizontalredlinerepresents theupperboundofthe95%confidenceintervalobtainedfromthesimulations.Intheinsets,wereportthedistributionofdMcomingfrom1000000
simulationsoftheunbiasedsamplingprocess.Again,theredlinerepresentstheupperboundofthe95%confidenceinterval.(Forinterpretationofthe referencestocolorinthisfigurelegend,thereaderisreferredtothewebversionofthearticle.)
Fig.6. Heatmapsshowingthebiasbyfieldandageoftherankingsbythedifferentindicators.Eachcellrepresentsanage-fieldgroup:agegroupsare representedhorizontally,whilefieldsarerepresentedvertically.Thecolorofthecellsshowsthebiasoftheindicatorswithrespecttothatage-fieldgroup. Whitemeansthattherespectiveage-fieldgroupisfairlyrepresentedinthetop1%oftherankingbytheindicator.Whileweuseacolorscalefromwhite tointensered(blue)forage-fieldgroupwhichareunderestimated(overestimated).(Forinterpretationofthereferencestocolorinthisfigurelegend,the readerisreferredtothewebversionofthearticle.)
theimprovedindicators,age-andfield-rescaledcitationcountandPageRank,wealsohavetorejectthenullhypothesis, eventhoughtheyaremuchclosertothe95%confidenceinterval.
Itisworthtonoticethatage-rescaledPageRank,anindicatordevelopedtoonlyremovePageRank’sbiasbyage(Mariani etal.,2016),issignificantlylessbiasedcomparedtorelativecitationcount,anindicatorspecificallydesignedto simultane-ouslyremovebiasbyageandfield.
5.4. Simultaneouslyvisualizingthebiasbyageandfield
Tovisualizethefieldandagebiasoftherankingsbytheanalyzedindicators,weuseheatmapsintheage-fieldgroupplane (seeFig.6).Intheseheatmaps,eachcellrepresentsafield-agegroup,anditscolorindicatesthelevelofbias.Awhitecell indicatesthatthenumberofpapersintherespectiveage-fieldgroupfallsintothe95%confidencelevel(C95%)determined
withthesimulations.Hence,whitemeansthatnobiasisdetectedfor thatage-fieldgroup.Whileforrepresentingthe biastowardsoragainstagroupofpapers,weuseblue(overestimation)andred(underestimation).Toobtainarangeof over/under-estimation,thebrightnessofthecolorsrangesfromwhite(nobias)tointenseblue/red.Themostintensecolors indicatethatthenumberofpapersfromthatage-fieldgroupis5standarddeviationsmaller/biggerthantheexpectedvalue. ThetoppanelofFig.6showsthat,independentlyoffield,citationcountandPageRanksystematicallyover-represent oldpapersandunder-representrecentpapers.Thisisinagreementwiththefindingsofseveralotherworks(Chenetal., 2007;Marianietal.,2015,2016;Newman,2009).TheonlyexceptionisPoliticalSciencewhichisusuallyunderestimated independentlyofpaperage.Wearguethatthishappensbecausethisisthesmallestfieldinthedataset,andithasbecome anacademicdisciplinebyitselfmuchlatercomparedtomostoftheotherfields3.Also,theoldestpapersinmostfieldsare
under-representedbycitationcount,whichreflectsthechangeofcitationpracticesovertime.
ThemiddlepanelsofFig.6showthattherelativecitationcountandage-rescaledPageRanksuppresslargepartofthe biasesoftheoriginalmetrics,yetspecificfieldsareconsistentlyoverestimatedorunderestimated.Forexample,both age-rescaledPageRankandrelativecitationcountunder-representpapersbelongingtothefieldofChemistry.Apeculiarityof relativecitationcountisthatitover-representsboththeoldestaswellasthemostrecentpapersatthecostoftheother papers.
ThebottompanelsofFig.6showtheheatmapsforthenewindicatorsage-andfield-rescaledcitationcountRAF(c)and
PageRankRAF(p).Wefindthattherespectiverankingsaremuchlessbiasedtowardsspecificfieldscomparedtoalltheother
analyzedmeasures.However,therearetwopatterns:forRAF(c)recentpublicationstendtobeunderestimatedforsome
fields,whereasforRAF(p)recentpublicationtendtobeoverestimatedforalmostallfields.Theserathersystematicpatterns
musthavetheirrootsinchangesofthecitationandPageRankscoredistributionswithtime.Sinceourrescalingprocedure wasfixingthefirsttwomomentsofthesedistributions,theobservedpatternscomefromdifferencesinhighermoments. Thus,thedistributionsofRAF(c)andRAF(p)arealignedonlypartiallyforpapersofdifferentage.
6. Discussionandconclusion
Tosummarize,inthispaperwehaveanalyzedalargecitationnetworkfromtheMicrosoftAcademicGraphtoshowthat therankingsofpapersbywell-knownindicatorsareextremelybiasedbyageandfield.Thelevelofbiasoftherankingshas beenquantifiedwithanewstatisticalframeworkbasedontheMahalanobisdistance.Thisframeworkhasallowedusto simultaneouslyquantifytheageandfieldbiasesoftheanalyzedrankings,andtodeterminewhichgroupsofpapersgivethe largestcontributionstotheobservedbias.Toallowotherresearchertoeasilyimplementourstatisticaltestforrankingbias, wemaketherespectivecodepubliclyavailable4togetherwithaquicktutorialonhowtouseit.5Inaddition,wehavealso
introducedtwonewindicatorsofpaperimpact,rescaledcitationcountRAF(c)andrescaledPageRankRAF(p)thatproduce
muchlessbiasedrankingsthanexistingindicators.Inparticular,therankingbyRAF(p)isapproximatelythreetimesless
biasedcomparedtotheleastbiasedexistingmetrics,relativecitationcountandage-rescaledPageRank.
Thecontributionofourresultstothedebateonthevalidityoffield-normalizationproceduresisthreefold.First,our findingsareinagreementwiththeconclusionsofAlbarránetal.(2011)andWaltmanetal.(2012)whicharguedthatthe relativecitationcountintroducedbyRadicchietal.(2008)canbeinsufficienttoeffectivelyremovecitationcount’sbiasby ageandfield.Second,weshowtheimportanceoftestingindicatorsusinganaccuratestatisticalprocedure,suchtheone introducedinthispaper.Indeed,fortheleast-biasedindicatorsanalyzed,RAF(c)andRAF(p),noclearindicationofbiasisfound
atfirstglance.However,whenusingthestatisticaltestbasedontheMahalanobisdistance,wefindasignificantdiscrepancy betweentheirrankingsandthosecomingfromunbiasedsamplingprocess.Wearguethatincludinghigher-ordermomenta (suchastheskewness)intherescalingprocedurecanbeanefficientwaytofurtherreducetherankings’levelofbias.Third, byderivinganexplicitformulatocalculatethecontributionofeachfieldtothebiasofaranking,wefindthatthethese contributionsassumeabroadrangeofvalues.Weobtainsimilarfindingsalsoforthecontributionstotheage-fieldbias.This meansthatthelevelofbiasofrankingsdependsheavilyonwhichyearsorfieldsareincludedintheanalysis.Forthisreason, infutureresearchonageandfieldnormalizationofindicators,itisessentialtoclearlymotivatewhichyearsandfieldsare includedintheanalysis,avoidingarbitraryoruncriticaldecisions.
Toaddressthebiasbyageandfieldofrankingofpapers,wehavefirstdividedthepapersingroupswithsimilarageand fromthesamefield.Then,weconsideredonlythesizesofthesegroupstodefineanunbiasedselectionprocessfromwhich weobtainedastatisticalnullmodelforanunbiasedranking.Inprinciple,additionalinformationcanbeincludedintothe nullmodeltocorrectforothereffects.Forexample,includinginformationabouttheco-authorshipnetworkwouldpermit tocorrectfortheeffectofthisnetworkonthegrowthandstructureofthecitationnetwork(Sarigöl,Pfitzner,Scholtes,Garas,
3We notice thatthe classification of PoliticalScience as oneof the highest-levelfields is not obvious. In Scopuscategories, “Political Sci-enceandInternationalRelations”isonlyasubfieldofthehigher-levelfieldSocialScience[http://www.scimagojr.com/journalrank.php?area=3300]. In the Web of Science classification scheme, “Political Science” is only a subfield of the higher-level field “Social Sciences, General” [http://ipscience-help.thomsonreuters.com/inCites2Live/8300-TRS.html].
4https://github.com/giava90/quantifying-ranking-bias.
5https://www.sg.ethz.ch/team/people/gvaccario/quantifying-ranking-bias/.
&Schweitzer,2014;Sarigol,Garcia,Scholtes,&Schweitzer,2017).Inthisway,wewouldgainabetterunderstandingofhow thesocialdimensionofsciencecontributestothefieldandagebiasesofimpactindicators.
Weemphasizethatremovingthebiasesaddressedinthispaperandthosethatcomefromsocialaspectsisofprimary importancenotonlyforscholarlypublicationdatabases,butalsoforseveralotherinformationsystems,suchastheWWW oronlinesocialnetworks(Scholtes,Pfitzner,&Schweitzer,2014a).Asamatteroffact,everydayscholarsandon-lineusers exploreavailableknowledgeusingrecommendersystemsbasedonrankingalgorithms.Thischallengesustodesignmore sophisticatedfilteringandrankingprocedurestoavoidbiasesthatcansystematicallyhiderelevantcontentsoronlyshow informationtoosimilartowhattheusersalreadyknow.
Whilewe haveanalyzedindetail thepresence ofageandfieldbiasin theranking,it stillremainstoevaluatethe actualrankingperformanceofthenewlyproposedindicatorsinartificialdata(Medo&Cimini,2016)orinrealdatawhere thegroundtruthisprovidedbysomeexternalsource(Dunaiski,Visser,&Geldenhuys,2016;Marianietal.,2016).Another importantissueisthecomparisonbetweenmetricsbasedoncitationcountandmetricsthattakethewholecitationnetwork intoaccounttodeterminepapers’score.Ouranalysis(seeSectionD)showsthattherankingsbytheleast-biasedindicators RAF(c)(citation-based)andRAF(p)(network-based)arepositivelycorrelated,stillsubstantiallydifferent.Canweusethe
extra-informationprovidedbythenetworktoenhanceourabilitytoidentifyhighly-significantpublications?Ourintuition andtheresultspresentedbyChenetal.(2007)andbyMarianietal.(2016)forspecificresearchfieldssuggestthatthisisthe case.Yet,weneedadditionalanalysistovalidatethisconjectureinlargerdatasetssuchastheoneanalyzedinthisarticle.
Toconclude,byreducing theageandfieldbiasesfromindicatorsofscientificimpactandbyextendingtheexisting statisticaltestsforbiases,wecontributetothechallengeofquantifyingandsuppressingbiasesofrankingsininformation systems.
Acknowledgements
WethankIngoScholtes,FrankSchweitzerandYi-ChengZhangforsuggestionswhichimprovedthemanuscript.In addi-tion,wealsothankEmreSarigölforhishelpinpre-processingthedata,andEliasBaumanforhisimportantcontribution totheoptimizationoftheC++codethatweusedforthenetworkanalysis.GVacknowledgessupportfromtheSwissState SecretariatforEducation,ResearchandInnovation(SERI),GrantNo.C14.0036aswellasfromEUCOSTActionTD1210 KNOWeSCAPE.MSMacknowledgessupportfromtheSwissNationalScienceFoundationGrantNo.200020-156188.
AuthorcontributionsConceivedanddesignedtheanalysis:GiacomoVaccario,MatusMedo,NicolasWider,Manuel
Sebas-tianMariani.
Collectedthedata:GiacomoVaccario.
Contributeddataoranalysistools:GiacomoVaccario,ManuelSebastianMariani. Performedtheanalysis:GiacomoVaccario,ManuelSebastianMariani.
Wrotethepaper:GiacomoVaccario,MatusMedo,NicolasWider,ManuelSebastianMariani.
AppendixA. KDDCupdata
A.1Datasource
Inthiswork,weanalyzedthedumpoftheMicrosoftAcademicGraph(MAG)releasedfortheKDDCup2016(Sinhaetal., 2015),acompetitionlinkedtoaprestigiouscomputerscienceconferenceonknowledgediscoveryanddatamining(KDD). AmongtheprimaryinterestsofthecommunityorganizingtheKDDCup,therearethetechnicalchallengesrelatedto web-scaledatacollectionandaggregation.Forthisreason,thereleaseddatafortheKDDCup2016wentthroughonlybasic processing.6Eachpublicationinthedatasetisendowedwithitsuniqueidentifier,paperid,publicationdate,referencesto
otherpublications,anditsfieldofstudy. A.2Datapre-processing
Whenanalyzingthedata,wedonotdistinguishthepublicationsbytheirtype(paper,review,book,etc.).Further,we alsodonotdifferentiatebetweendifferenttypesofjournalsandtakeintoaccountallofthem:forexample,wedonot distinguishbetweenacitationcoming(orgoing)from(orto)aletterorabook.Wearguethatitisimportanttokeepvarious typesofjournalsandpublicationsbecausedifferentfieldsadoptnotonlydifferentcitationnorms,butalsodifferentwaysto communicatetheirresults.Forexample,computerscienceresearcherscommonlypublishresultsinconferenceproceedings, whilephysicsauthorstendtopreferarticlesorletters.Atthesametime,weareawarethatdifferenttypesofpublications mighthavedifferentcitationcharacteristics.However,goodindicatorsshouldideallybeabletoaccountforheterogeneities amongpublicationsandcitationnormsacrossdifferentcommunitiesandproduceunbiasedrankingswithouttheneedfor
6https://kddcup2016.azurewebsites.net/Data
TableA.2
Mainfields.The19parentmainfieldsidentifiedbyMicrosoftAcademicGraphwiththeirnumberofpublicationsandaveragecitations.Thesecondtolast rowreportsthetotalnumberofpublicationsconsideringmultipletimespublicationsthatbelongtomorethanonefields,whereasthelastrowreportsthe totalnumberofuniquepublicationsandtheaveragecitationcount.
Field Publicationcount Meancitationcount
Art 233251 3.90 Biology 5847554 9.67 Business 613827 4.81 Chemistry 6204531 7.28 ComputerScience 4080636 6.13 Economics 2252921 5.56 Engineering 3011763 5.10 EnvironmentalScience 315465 12.63 Geography 288338 6.92 Geology 1825707 7.88 History 390144 5.53 MaterialsScience 2063474 6.12 Mathematics 4551453 5.87 Medicine 5061990 7.90 Philosophy 787649 5.05 Physics 6976644 5.55 PoliticalScience 144473 2.51 Psychology 2861813 8.23 Sociology 1784695 5.39 Total 49296327 –
Total(nomultiple) 18193082 6.42
arbitrarychoicesaboutwhichtypesofarticlestoincludeintheanalysis.Inaddition,similarlyasWaltmanetal.(2012)and differentlyfromRadicchietal.(2008),wedonotexcludepublicationswhichdonotreceivecitations.
AsmentionedinSection2,theMAGhasafieldclassificationschemewith4hierarchicallevels.Thefieldassignmentis basedonaninternalalgorithmthatusesamachinelearningapproach(Sinhaetal.,2015).Inourwork,similarlytoRadicchi etal.(2008),weareonlyinterestedinimpactmetricnormalizationatthemostcoarse-grainedlevel.Tothisaim,inour analysis,wefocusonlyonthe19mainfieldsaslistedinTableA.2.Discussingthepossiblelimitationsoftheclassification approachbyMASandthedependenceofourresultsontheadoptedclassificationschemeisarelevantsubjectforfuture researchbutgoesbeyondthescopeofthismanuscript.Weonlyincludedintheanalysispapersforwhichthefollowing informationareavailable:(1)uniqueidentifier(ID);(2)completepublicationdate(yyyy/mm/dd)crucialforthetemporal rescalingprocedurethatisexplainedinthefollowing;(3)DOIorjournal-id,inordertobeabletoretrievethepublication; (4)assignmenttoatleastoneofthemain19fields.Wediscardfromouranalysispublicationsforwhichoneormoreof thesefourpropertiesaremissing.
Withthisfilteringprocedure,weobtainN=18193082uniquepublicationsandE=109719182citations. A.3Databasicstatistics
Wefirstobservethatthedistributionsofbothincomingandoutgoingedgesarebroad(seeFig.A.7).Boththesedistribution infacthavelongandheavytails.Inaddition,wealsoobservestrongvariationsofthemeancitationcountacrossfields(see TableA.2fordetails).Forexample,publicationsbelongingtothefieldofEnvironmentalScienceandofPoliticalSciencehave
Fig.A.7.In-andout-degreedistributionforthepublicationspresentinourdatasetafterpreprocessingthedata.
Fig.A.8.ScatterplotofthecitationcountsreportedinthedatareleasedfortheKDDCup2016andfromtheonlineversionoftheMAG(02/2017).
meancitationcounttwotimesbiggerandsmaller,respectively,comparedtotheaveragecitationcountacrossallfields.In agreementwiththefindingsbyRadicchietal.(2008),SchubertandBraun(1986),Vinkler(1986),thestrongvarietyinmean citationcountperpublicationforthevariousfieldsconfirmsthatthecorrespondingcommunitiesexhibitdifferentcitation behavior,whichcallsforfieldnormalizationprocedures.
A.4Dataquality
Here,weaddressthefollowingquestion:towhichextentistheKDDdumpoftheMAGinaccordancewiththemost updatedonlineversionoftheMAG?7Forthiscomparison,wedividedthe49296327analyzedpaper-fieldpairs(one
paper-fieldpairiscomposedofapaperandthefieldsthatpaperbelongsto)into760ageandfieldgroups,asdescribedinSection5. Fromeachgroup,werandomlychoose0.1%ofpapers.Followingthisprocedure,weobtainedarepresentativesamplewith respecttofieldandagecomposedofapproximately50000papers.
First,wehavematchedthepaperIDinhexadecimalformatpresentinthedatareleasedfortheKDDCuptothepaper IDinint64formatpresentintheon-lineversionoftheMAG.For50papers,wemanuallyverifiedthatthepaperIDswere exactlythesameinthetwodatasets,albeitrepresentedindifferentformats.Then,usingtheAcademicKnowledgeApi,8we
havedownloadedthenumberofcitationsforeachsampledpaper.
Forthe2.3%ofthesampledpapers,wewerenotabletofindtheircorrespondingpaperintheonlineMAG. For50 unmatchedpapers,wefoundoutthatthesewerepaperswithduplicatesintheKDDCupdata,i.e.thesepapersarepresent intheKDDdatawithtwodistinctIDs,andonlyoneofthetwoIDsispresentintheMAG.
Forthematchedpapers(97.7%ofthesample),wecomparedthenumberofcitationsreportedintheKDDdumpofthe MAGwiththenumberofcitationsreportedontheonlineversionoftheMAG.SincetheonlineMAGhasalongertimespan thantheKDDdata,intheabsenceofnoise,wewouldexpectthenumberofcitationsintheKDDdatatobesmallerthan orequaltothenumberofcitationsreportedintheonlineMAG.Wefindthatthetwocitationcountsarehighlycorrelated (Fig.A.8),andonlythe2.2%ofthesampledpapershavemorecitationsintheKDDdatacomparedtotheonlineversionof theMAG.Thissetsalowerboundaryfortheerrorinthepercentageofpaperswithwrongnumberofcitationstoabout2.2%. Tosummarize,afterourfilteringprocedure,wefindthatthedatareleasedforKDDCup2016hasabout2.3%ofpapers withduplicates.Inaddition,about2.2%ofthematchedpapershaveerrorsintheircitationcount.Thismeansthatwehave correctcitationinformationforabout95.6%oftheanalyzedpapers.
AppendixB. EvaluatingindividualcontributionstotheMahalanobisdistance
TheMahalanobisdistance(dM)isanestablishedmeasureinstatisticswhichgeneralizestheconceptofz-scoreto multi-variatedistributionsbytakingintoaccountalsopossiblecorrelationsbetweentherandomvariables(Mahalanobis,1936). Itsdefinitionreads
dM(x, y)=
(x − y)TS−1(x − y) (B.1)
7WehaveperformedthisanalysisduringFebruary2017.
8https://www.microsoft.com/cognitive-services/en-us/academic-knowledge-api
whereS−1 istheinverseofthecovariancematrix,x and y aretwovectorscontainingtherandomvariables.Whenthe
covariancematrixisdiagonal,i.e.therandomvariablesarenotcorrelated,thanthedMisequivalenttothesquarerootof thesumofthesquaresofthez-scores.
InSection5.2,wehaveusedEq.(7),anexpressionforthedMvalidwhenthecovariancematrixcomesfromaMultivariate HypergeometricDistribution(MHD),i.e.,whentheelementsofthematrixare
Sij=
ıij(Ki(N−Ki))− 1−ıij KiKj∀
i,j=1,...,F−1 (B.2)whereıijistheKroneckerdelta,Kiisthenumberofpapersofcategoryi,N=
FiKiisthetotalnumberofpapers,Fisthe
numberofpapercategories,=n(N−n)/N2(N−1)andnisthenumberofsampledpapers.Itisworthytorememberthat
eventhoughwehaveFdifferentcategories,weonlyhaveF−1degreesoffreedom.Here,wederiveEq.(7)forthecaseofa MHDinthreedimensions,i.e.,forF=3.Inthiscase,thecovariancematrixis2×2:
S=
K1(N−K1) −K1K2 −K1K2 K2(N−K2) (B.3) andtheinverseofthecovariancematrixisS−1= 1 det(S)
K2(N−K2) K1K2 K1K2 K1(N−K1) (B.4)wheredet(S)=K1(N−K1)K2(N−K2)−(K1K2)2denotesthedeterminantofthecovariancematrix,S.Then,letusconsidertwo
randomcolumnvectorsextractedfroma3-dimensionalMHD,x=(x1,x2,x3)Tandy=(y1,y2,y3)Tsuchthatn=
3i=1xi= 3i=1yiwherenisthenumberofsampledpapers.SubstitutingEq.(B.4)inEq.(B.1),wewritethesquareofthedMbetween xandyas dM(x, y)2= 1 det(S)
x1−y1 x2−y2K2(N−K2) +K1K2 +K1K2 K1(N−K1)
x1−y1 x2−y2
= 1 det(S) (x1−y1)2K2(N−K2)+(x2−y2)2K1(N−K1) +2(x1−y1)(x2−y2)(K1K2)) = det(S)1 (x1−y1)2K2(K1+K3)+(x2−y2)2K1(K2+K3) +2(x1−y1)(x2−y2)(K1K2) = 1 det(S) (x1−y1)2K2K3+(x2−y2)2K1K3 + [(x1−y1)+(x2−y2)]2K1K2
wherewehaveusedN=
3i=1Ki.Recallingthatn= 3i=1xi=
3i=1yi,weknowthat(x1−y1)+(x2−y2)=(x3−y3),sowe
write
dM(x, y)2= det(S)1
(x1−y1)2K2K3+(x2−y2)2K1K3+(x3−y3)2K1K2(B.5) Then,byusingtherelationdet(S)=N
3i=1Ki,wehave:dM(x, y)2=1 3
i=1 (xi−yi)2 NKi ; (B.6)noticingfromEq.(B.2)that=Si,i/(Ki(N−Ki)),weobtain
dM(x, y)2= 3
i=1 (xi−yi)2 Sii Ki(N−Ki) NKi = 3 i=1 (xi−yi)2 Sii 1−Ki N (B.7)Finally,ifwechooseoneofthetwovectorstocontaintheexpectedvalues,i,were-obtainEq.(7)since(xi−i)2/Sii=zi2.
Tobeprecise,thecovariancematrixisnotdefinedfori=3,howevertherelation=2
3/(K3(N−K3))holdsandtherefore
alsothefinalresult.
UsingMathematicaorsimilarsoftwares,itiseasytoproveanalyticallythateq.(7)holdsforsmalldimensions.Wehave verifiedituntil6dimensions.Moreover,wehavenumericallytestedthisformulabycalculatingthedM’sbetweentheranking vectorsoftheindicatorsandthevectorofexpectedvalues, ,withtwodifferentalternativemethods:(1)byusingEq.(B.1),
Fig.C.9.Distributionof2obtainedbycalculatingtheempiricalratio,r=(e2
−1),betweenthevarianceandthesquareofthemeancitationcountof eachfieldandyear.
i.e.,byinvertingthecovariancematrix,and(2)byusingtheeigenvaluedecompositionofthecovariancematrix.9Theresults
ofthethreemethodswereallcompatiblewitheachotherupto10decimaldigits.TheadvantageofusingEq.(7)isthatwe cancalculatethedMbetweentwoarbitraryvectorswithoutdealingwithany(computationallyslow)matrixinversionor diagonalization,andthenumberofneededcalculationsscaleslinearlywiththenumberofdimensions.Importantly,Eq.(7) allowsalsotoassesstheindividualcontributionofeachdimension(i.e.ofeachcategory)tothed2
M.Toourbestknowledge, wearethefirstonestohavederivedsuchexplicitformulafordMwhenthecovariancematrixandtherandomvectorscome fromaMHD.
AppendixC. Firstmomentrescaling
AccordingtoRadicchietal.(2008),thedistributionofthecfindicatorislog-normal:
F(cf)dcf = 1
cf√2e
−[log(cf)−]2/22
dcf (C.1)
where=−2/2andisfittedfromthedata.WhenEq.(C.1)isverified,thenalsothedistributionsofcitationcount,c i,for
alltheindividualfields,i,arelognormal: F(ci)dci=
1 ci√2e
−[log(ci)−log(c0)−]2/22dc
i (C.2)
wherec0isthemeanofci.Forlognormaldistributionsthevarianceisproportionaltothesquareofthemeanandthe
constantofproportionalityis(e2
−1).FromEq.(C.2),weseethatthecitationcountsciaredistributedlognormallywith
meane+log(c0)+2/2 andvariance(e2−1)e2+2log(c0)+2.Recallingthat=−2/2,wehavethatthemeanisc0,asitis expected,whilethevariancebecomes(e2
−1)c2
0.Thus,wheneq.(C.1)isverified,thevarianceoftheempiricaldistribution
ofthecitationsforeachfieldhastobeproportionaltothesquareofthemeancitationcount.Moreover,theconstantof proportionalityhastobe(e2
−1)foreveryfieldandyear.
TheanalyticresultjustpresentedisinlinewithEq.(C.1)giveninAppendixCofMarianietal.(2016).Thereitisshown thatarescalingprocedurebasedondivingtheoriginalscorebytheirfirstmomentworksiftheratiobetweenstandard deviationandmeanisconstant.Inthecaseoftherelativecitationratio,wecancalculateanalyticallysuchconstantusing thelognormaldistributionandobtainthefittingparameter2.
InFig.C.9,wereportthedistributionof2obtainedbycalculatingtheempiricalratiobetweenthevarianceandthe
squareofthemean,r,andbyinvertingtherelationr=(e2
−1)foreveryfieldandyear.Iftheuniversalityclaimwascorrect, wewouldexpectanarrowdistributionof2.Bycontrast,wefindthat2rangesbetween0and8acrossdifferentfieldsand
years.Wearguethatthebroadrangeof2isthereasonwhythefirstmomentrescalingintroducedin(Radicchietal.,2008)
doesnotworkintheanalyzeddataset.
9ThematrixSissymmetricandithasmaximalrankbecauseitisthecovariancematrixofamultivariatedistribution.Therefore,wecandiagonalizeit, S=B−1DBwherethecolumnsofBformanorthonormalbasis;wecanalsowriteS−1=B−1D−1B.Withthis,wehaved2
M(x, y)=
F−1i ci/i,where{i}are theeigenvaluesofSand{ci}arethecoordinatesofx − y inthebasiswhichdiagonalizesS,i.e.ci=
F−1 k (xk−yk)B−1 ki =
F−1k (xk−yk)Bikwherethelast equalitycomesfromtheorthonormalityofBwhichimpliesB−1=BT.
TableD.3
Correlationsbetweenthemetrics.Thefourcolumnsrepresent(fromlefttoright):Pearson’scorrelationcoefficientr,rrestrictedtopapersthatreceivedat leasttencitations,Spearman’srankcorrelationcoefficient ,and restrictedtopapersthatreceivedatleasttencitations.
Metrics r r(onlyc>10) (onlyc>10)
c,p 0.82 0.82 0.79 0.59
RAF(c),RAF(p) 0.80 0.81 0.82 0.74
AppendixD. Comparingtherankingsbythemetrics
InSection5,wehaveshownthattherankingsbycitationcountandPageRankcanbesubstantiallyleveledoffacross differentfieldsandagegroupsthrougharescalingprocedurebasedonthez-score.Thetworesultingindicators,RAF(p)
andRAF(c),aretheleastbiasedamongtheindicatorsconsideredinthispaper.Itisimportanttonoticethatwhilecitation
counthasadirectinterpretationanditiswidelyusedinresearchevaluation(Waltman,2016),PageRankscoreisamore sophisticatedquantitywhichhasnotyetbeturnedintoastandardtoolforresearchassessment.Iftherankingsbyrescaled PageRankandrescaledcitationcountbroughtsimilarinformation,rescaledcitationcountmightbepreferredduetoits simplerinterpretationandeasiercomputation.
Differentlyfromthecitationcount,PageRankscoreusesinformationonthewholenetworktopologytocomputeeach paper’sscore.WhilethereexistsnouniversalcriteriontodecidewhetherPageRankscoreleadstoabetterrankingthanthe citationcount,theresultsbyChenetal.(2007)andMarianietal.(2016)suggestthatincitationnetworks,PageRankscore improvesourabilitytofindgroundbreakingpublicationsinthedata.Ontheotherhand,anodethatreceivedmanycitations ismorelikelytoachievelargerPageRankscore–Fortunato,Boguná,Flammini,andMenczer(2006)showedthatPageRank scoreisonaverageproportionaltocitationcountforuncorrelatednetworks.
Weaddressnowthequestion:Towhatextenttherankingsby(rescaled)PageRankand(rescaled)citationcountdiffer?We focusontwocorrelations:(1)thecorrelationbetweencitationcountandPageRank,whichhasbeenofinterestforprevious studies(Chen,Gan,andSuel(2004),Fortunatoetal.(2006),Pandurangan,Raghavan,andUpfal(2002))duetotheessential roleplayedbyPageRankalgorithmindeterminingthesuccessofGoogle’sWebsearchengine;(2)thecorrelationbetween thetworescaledindicatorsRAF(c)andRAF(p).Themeasuredcorrelationsareallpositive,significantlylargerthanzero,yet
significantlysmallerthanone(seeTableD.3).Thisisinterestingasitmaypointoutthat,inanalogywiththefindingsby Chenetal.(2007),Marianietal.(2016),Ren,Mariani,Zhang,andMedo(2017),networktopologybringsusefulinformation thatisneglectedbycitationcount.WhethertheadditionalinformationusedbyPageRankmetriccanbeusedtoidentify groupsofsignificantpaperswillbethesubjectoffutureresearch.
DifferentlythantherankingsbycitationcountandPageRankscore(notshownhere),thetoppapersidentifiedbyRAF(c)
andRAF(p)comefromdiversehistoricalperiodanddiversefields.Duetothelackoftimebias,alsoveryrecentpaperscan
reachthetopoftherankingsbyRAF(c)andRAF(p).Itwouldbeinstructivetolookatthetoppapersasrankedbyrescaled
citationcountandrescaledPageRank.However,theoriginalMAGdatasetspresentsnoisyentries,asreportedbytheKDD cup2016(seeAppendixAformoredetails),anditcausesthescoresofsomerecentpaperstobeover-estimated,which makessomeoftheentriesinthetop-20oftherankingsbytherescaledscoresunreliable.Forthisreason,wedonotshow therankingshere.Atthesame,thisproblemdoesnotaffectthestatisticalresultspresentedinthepreviousSections.
References
Adams,J.,Gurney,K.,&Jackson,L.(2008).Calibratingthezoomatestofzittshypothesis.Scientometrics,75(1),81–95.
Albarrán,P.,Crespo,J.A.,Ortuno,I.,&Ruiz-Castillo,J.(2011).Theskewnessofsciencein219sub-fieldsandanumberofaggregates.Scientometrics,88(2), 385–397.
Bergstrom,C.T.,West,J.D.,&Wiseman,M.A.(2008).Theeigenfactormetrics.JournalofNeuroscience,28(45),11433–11434. Bonacich,P.(1987).Powerandcentrality:Afamilyofmeasures.AmericanJournalofSociology,92(5),1170–1182.
Bornmann,L.,&Daniel,H.-D.(2008).Whatdocitationcountsmeasure?Areviewofstudiesoncitingbehavior.JournalofDocumentation,64(1),45–80. Bornmann,L.,&Daniel,H.-D.(2009).Universalityofcitationdistributions–Avalidationofradicchietal.’srelativeindicatorcf=c/c0atthemicrolevel
usingdatafromchemistry.JournaloftheAmericanSocietyforInformationScienceandTechnology,60(8),1664–1670.
Brin,S.,&Page,L.(1998).Theanatomyofalarge-scalehypertextualwebsearchengine.ComputerNetworksandISDNSystems,30(1),107–117. Chen,Y.-Y.,Gan,Q.,&Suel,T.(2004).Localmethodsforestimatingpagerankvalues.ProceedingsofthethirteenthACMinternationalconferenceon
informationandknowledgemanagement,ACM,381–389.
Chen,P.,Xie,H.,Maslov,S.,&Redner,S.(2007).FindingscientificgemswithGoogle’sPageRankalgorithm.JournalofInformetrics,1(1),8–15.
Colliander,C.,&Ahlgren,P.(2011).Theeffectsandtheirstabilityoffieldnormalizationbaselineonrelativeperformancewithrespecttocitationimpact: Acasestudyof20naturalsciencedepartments.JournalofInformetrics,5(1),101–113.
deSollaPrice,D.(1976).Ageneraltheoryofbibliometricandothercumulativeadvantageprocesses.JournaloftheAmericanSocietyforInformationScience, 27(5),292–306.
Dunaiski,M.,Visser,W.,&Geldenhuys,J.(2016).Evaluatingpaperandauthorrankingalgorithmsusingimpactandcontributionawards.Journalof Informetrics,10(2),392–407.
Ermann,L.,Frahm,K.M.,&Shepelyansky,D.L.(2015).Googlematrixanalysisofdirectednetworks.ReviewsofModernPhysics,87(4),1261. Fortunato,S.,Boguná,M.,Flammini,A.,&Menczer,F.(2006).ApproximatingPageRankfromin-degree.InInternationalworkshoponalgorithmsand
modelsfortheweb-graph.pp.59–71.Springer.
Franceschet,M.(2011).Pagerank:Standingontheshouldersofgiants.CommunicationsoftheACM,54(6),92–101. Garfield,E.(2006).Thehistoryandmeaningofthejournalimpactfactor.JAMA,295(1),90–93.
Gleich,D.F.(2015).Pagerankbeyondtheweb.SIAMReview,57(3),321–363.
Harzing,A.-W.,&Alakangas,S.(2013).Microsoftacademic:Isthephoenixgettingwings?Scientometrics,1–13.
Hirsch,J.E.(2005).Anindextoquantifyanindividual’sscientificresearchoutput.ProceedingsoftheNationalacademyofSciences,1656,9–16572. Hug,S.E.,&Braendle,M.P.(2017).ThecoverageofMicrosoftacademic:Analyzingthepublicationoutputofauniversity.,arXivpreprintarXiv:1703.05539 Hug,S.,Ochsner,M.,&Brandle,M.P.(2017).CitationanalysiswithMicrosoftacademic.Scientometrics,111(1),371–378.
Katz,L.(1953).Anewstatusindexderivedfromsociometricanalysis.Psychometrika,18(1),39–43.
Kleinberg,J.M.(1999).Authoritativesourcesinahyperlinkedenvironment.JournaloftheACM(JACM),46(5),604–632. Lundberg,J.(2007).Liftingthecrowncitationz-score.JournalofInformetrics,1(2),145–154.
Mahalanobis,P.C.(1936).Onthegeneraliseddistanceinstatistics.ProceedingsoftheNationalInstituteofScienceofIndia,4,9–55. Mariani,M.S.,Medo,M.,&Zhang,Y.-C.(2015).Rankingnodesingrowingnetworks:WhenPageRankfails.ScientificReports,5.
Mariani,M.S.,Medo,M.,&Zhang,Y.-C.(2016).Identificationofmilestonepapersthroughtime-balancednetworkcentrality.JournalofInformetrics,10(4), 1207–1223.
Maslov,S.,&Redner,S.(2008).PromiseandpitfallsofextendingGoogle’sPageRankalgorithmtocitationnetworks.JournalofNeuroscience,28(44), 11103–11105.
McAllister,P.R.,Narin,F.,&Corrigan,J.G.(1983).Programmaticevaluationandcomparisonbasedonstandardizedcitationscores.IEEETransactionson EngineeringManagement,4,205–211.
Medo,M.,&Cimini,G.(2016).Model-basedevaluationofscientificimpactindicators.PhysicalReviewE,94(3),032312. Medo,M.,Cimini,G.,&Gualdi,S.(2011).Temporaleffectsinthegrowthofnetworks.PhysicalReviewLetters,107,238701. Newman,M.E.J.(2009).Thefirst-moveradvantageinscientificpublication.EPL(EurophysicsLetters),86(6),68001.
Pandurangan,G.,Raghavan,P.,&Upfal,E.(2002).UsingPageRanktocharacterizewebstructure.InInternationalcomputingandcombinatoricsconference. pp.330–339.Springer.
Parolo,P.D.B.,Pan,R.K.,Ghosh,R.,Huberman,B.A.,Kaski,K.,&Fortunato,S.(2015).Attentiondecayinscience.JournalofInformetrics,9(4),734–745. Pinski,G.,&Narin,F.(1976).Citationinfluenceforjournalaggregatesofscientificpublications:Theory,withapplicationtotheliteratureofphysics.
InformationProcessing&Management,12(5),297–312.
Radicchi,F.,&Castellano,C.(2011).Rescalingcitationsofpublicationsinphysics.PhysicalReviewE,83(4),046116.
Radicchi,F.,&Castellano,C.(2012a).WhySirtes’sclaims(Sirtes,2012)donotsquarewithreality.JournalofInformetrics,6(4),615–618.
Radicchi,F.,&Castellano,C.(2012b).Testingthefairnessofcitationindicatorsforcomparisonacrossscientificdomains:Thecaseoffractionalcitation counts.JournalofInformetrics,6(1),121–130.
Radicchi,F.,Fortunato,S.,&Castellano,C.(2008).Universalityofcitationdistributions:Towardanobjectivemeasureofscientificimpact.Proceedingsof theNationalAcademyofSciences,105(45),17268–17272.
Ren,Z.-M.,Mariani,M.S.,Zhang,Y.-C.,&Medo,M.(2017).Atime-respectingnullmodeltoexplorethestructureofgrowingnetworks.,arXivpreprint arXiv:1703.07656.
Sarigöl,E.,Pfitzner,R.,Scholtes,I.,Garas,A.,&Schweitzer,F.(2014).Predictingscientificsuccessbasedoncoauthorshipnetworks.EPJDataScience,3(1),1. Sarigol,E.,Garcia,D.,Scholtes,I.,&Schweitzer,F.(2017).Quantifyingtheeffectofeditor-authorrelationsonmanuscripthandlingtimes.Scientometrics,
1–25(inpress).
Scholtes,I.,Pfitzner,R.,&Schweitzer,F.(2014).Thesocialdimensionofinformationranking:Adiscussionofresearchchallengesandapproaches.In Socioinformatics–ThesocialimpactofinteractionsbetweenhumansandIT.pp.45–61.Springer.
Scholtes,I.,Wider,N.,Pfitzner,R.,Garas,A.,Tessone,C.J.,&Schweitzer,F.(2014).Causality-drivenslow-downandspeed-upofdiffusionin non-Markoviantemporalnetworks.NatureCommunications,5.
Schubert,A.,&Braun,T.(1986).Relativeindicatorsandrelationalchartsforcomparativeassessmentofpublicationoutputandcitationimpact. Scientometrics,9(5-6),281–291.
Sinha,A.,Shen,Z.,Song,Y.,Ma,H.,Eide,D.,Hsu,B.-J.P.,etal.(2015).AnoverviewofMicrosoftacademicservice(MAS)andapplications.Proceedingsofthe 24thinternationalconferenceonworldwideweb,ACM,243–246.
Sirtes,D.(2012).FindingtheEastereggshiddenbyoneself:WhyRadicchiandCastellano’s(2012)fairnesstestforcitationindicatorsisnotfair.Journalof Informetrics,6(3),448–450.
vanLeeuwen,T.N.,&Medina,C.C.(2012).Redefiningthefieldofeconomics:Improvingfieldnormalizationfortheapplicationofbibliometrictechniques inthefieldofeconomics.ResearchEvaluation,21(1),61–70.
Vinkler,P.(1986).Evaluationofsomemethodsfortherelativeassessmentofscientificpublications.Scientometrics,10(3-4),157–177.
Walker,D.,Xie,H.,Yan,K.-K.,&Maslov,S.(2007).Rankingscientificpublicationsusingamodelofnetworktraffic.JournalofStatisticalMechanics:Theory andExperiment,2007(06),P06010.
Waltman,L.,&vanEck,N.J.(2010).Therelationbetweeneigenfactor,audiencefactor,andinfluenceweight.JournaloftheAmericanSocietyforInformation ScienceandTechnology,61(7),1476–1486.
Waltman,L.,Yan,E.,&vanEck,N.J.(2011).Arecursivefield-normalizedbibliometricperformanceindicator:Anapplicationtothefieldoflibraryand informationscience.Scientometrics,89(1),301–314.
Waltman,L.,vanEck,N.J.,&vanRaan,A.F.J.(2012).Universalityofcitationdistributionsrevisited.JournaloftheAmericanSocietyforInformationScience andTechnology,63(1),72–77.
Waltman,L.(2016).Areviewoftheliteratureoncitationimpactindicators.JournalofInformetrics,10(2),365–391.
Yao,L.,Wei,T.,Zeng,A.,Fan,Y.,&Di,Z.(2014).Rankingscientificpublications:Theeffectofnonlinearity.ScientificReports,4,6663.
Zhang,Z.,Cheng,Y.,&Liu,N.C.(2014).Comparisonoftheeffectofmean-basedmethodandz-scoreforfieldnormalizationofcitationsatthelevelofweb ofsciencesubjectcategories.Scientometrics,101(3),1679–1693.
Zhou,J.,Zeng,A.,Fan,Y.,&Di,Z.(2016).Rankingscientificpublicationswithsimilarity-preferentialmechanism.Scientometrics,106(2),805–816. Zitt,M.,Ramanana-Rahary,S.,&Bassecoulard,E.(2005).Relativityofcitationperformanceandexcellencemeasures:Fromcross-fieldtocross-scale
effectsoffield-normalisation.Scientometrics,63(2),373–401.