Graph drawing is a basic visualization tool. For graphs of up to hundreds of nodes and edges, there are many effective techniques available. At greater scale, data density and occlusion problems often negate its effectiveness. Conventional pan-and-zoom, an
TopologicalFisheyeViewsforVisualizingLargeGraphs
EmdenGansner
AT&TLabs—Research
YehudaKoren
AT&TLabs—Research
StephenNorth
AT&TLabs—Research
ABSTRACT
Graphhundredsdrawingisabasicvisualizationtool.Forgraphsofuptoavailable.oftenAtnodesgreaterandscale,edges,datatherearemanyeffectivetechniquestiscalenegatedensityandocclusionproblemsof-tionszoomingtoanditseffectiveness.Conventionalpan-and-zoom,andmul-thisgeometricproblem. sheyeAsanalternative,viewsarenotfullysatisfactorysolu-ofmethod.Itisbasedontheprecomputationwedescribeofatopologicalawithcoarsenedfoci.thelevelgraphs,ofdetailwhicharecombinedon-the- yintorenderingshierarchytechniqueWealsotoachievediscussdependentconstantarelatedinformationdistortiononthedistancefromoneormoredensitymethoddisplays.thatallowsourCRGraphics—ComputationalCategories:I.3.5[Computer[InformationUserInterfaces
Systems]:InformationGeometryMethodologies]:InterfacesandObjectandModelingComputerPresentation—H.5.2Keywords:topological sheye,largegraphvisualization1DrawingINTRODUCTION
nique.graphsisanerarchical”),Commonlayoutstylesimportantincludeinformationforcedirected,visualizationk-layeredtech-(“hi-onorthogonalandcircular[10,5].Inthispaper,wefocusiousForforcedirectedlayouts.
viewedlayoutgraphspanwithmethodsofmodestsize,withdozensofnodes,therearevar-ordinaryavailable.documentInviewersaddition,orgraphtheseviewerslayoutscanbelemsandputingarise.zoomdeveloped.thelayout.Thecontrols. rstAsgraphsbecomelarger,however,havingprob-Inproblemrecentisyears,thealgorithmicef cientalgorithmscomplexityofcom-andtechniquesUsefulfromapproacheslinearincludevirtualphysicalmodelshavebeen[5]PracticalGiventhegraphlayout,algebravisualcomplexityandstatisticsremains[8,13].
aproblem.Evenwithdif culttextatthedatalabelslowsetsendcanhavethousandsorevenmillionsofobjects.ofallofthatscale,itisnotrealistictoexpectlayoutssuchexceedasafortree.humansAttheunlesstheobjectshightheend,graphtobereadable,andnavigationisthenumberhassomeofspecialstructure,arethenumberofscreenpixels.Clearly,additionalobjectstechniquescanevenplorationHereneeded.
wedisplayedofconsiderlargegraphs.theproblemWeofthedisplayandinteractiveex-sentialobjectswhilepreservingseekastructuralwayofreducinginformationthenumberoftechniquetounderstandingthegraph.Inparticular,wepresentthataisnewes-(1)in-contexttheef cientforbrowsinguseofavailablelargegraphsdisplaywhosearea;characteristicsinclude:preservingdisplays;thegraph’sandstructure.
(3)avariabledegree(2)ofinformativeabstractiondetail-while2DistortionRELATEDproblemsviewingWORK
isanobviouscandidatefor sheyeSarkaranddistortionofbrowsingBrownhavelargegraphs.Variantsofdealing2-Dgeometricwiththe[20],beenandCarpendaledescribedby[2].LampingThesedoandagoodRaojob[14],of
e-mail:{erg,yehuda,north}@research.att.com
IEEE Symposium on Information Visualization 2004 October 10-12, Austin, Texas, USA 0-7803-8779-1/04/$20.00 ©2004 IEEE
highlightingobscuretransformation.
sinceaitfocusispackedarea,intobutmakeamuchthesmallerglobalareastructurebytheevennonlinearmoreofWeillustratethismethodusingthe4eltgraph[22].Thisgraph, ow15,606inaroundnodesa4andelement45,878edges,isameshcreatedtostudy uidKamada-Kawainodaldensity[23].Adrawingairfoil.ofItthisexhibitsgraphextremecomputedvariationsbytheinproperlyonitsrightcomesveryenlarged,handmethodcrowdedatside[12]isshowninFig.1.InFig.2,wezoomand

theusingobscure.
expensea sheyeoftherestlensof[2].theThisgraphportionthatbe-isFigure1:The4eltgraph,|V|=15,606,|E|

=45,878.
Figure2:A sheyeviewofthe4eltgraphfocusedonitsrighthandsideportion.
perbolicAnotherframe-rateprojectionapproach,inanproposedinteractivebyMunzner,viewerwithemployshighguaranteed3-Dhy-hugeitedtreesrenderingwitha3-D[17]. sheyeThiseffect.yieldsDisplayhighlyscalablecomplexityviewingofintendedbytreeof sheyethegraphfordepthgeneralandthataregraphs,renderednodesize.Thetechniqueisislim-notdistantfromanditsthelimitationsfocusareinsimilarobscuringtoareas[1,Multiscaleapproach.
the2-Dgation9,18].ofislessSuchgraph exiblelayoutsabstractionhasalsobeenconsidered;see,e.g.,dueshowtothetheexplicitglobalexpansionstructureandwell,contractionbutnavi-externallyclustersmaynotbeprovided.and,forsomeeasytosee.Also,methods,IntheFig. ne3westructuretheclustersshowthreeacrossthemselvesmustbedifferentdistinctlevelsclustersof
175
43, No. 3 COMMUNICATIONS OF THE ACM PERCEPTUAL USER INTERFACES Matthew Turk and George Robertson, Guest Editors computers by typing, pointing, and clicking...
Perceptual User Interfaces Matthew Turk Microsoft Research One Microsoft Way, Redmond, WA 98052 USA mturk@microsoft.com Abstract For some time, graphical ...
4页 免费 User Interfaces—Graphic... 暂无评价 2页 免费 Semantic User Interfaces 暂无评价 10页 免费 User Interfaces – Graph... 暂无评价 3页 免费...
CR Categories: H.5.1 [Information Interfaces and Presentation]: User Interfaces—Graphical user interfaces; Keywords: Human-Robot Interaction, User Interface ...
This dissertation defines and explores Graspable User Interfaces, an evolution of the input mechanisms used in graphical user interfaces (GUIs). A Graspable...
Presentation] User Interfaces—evaluation methodology, Preference-based search is the problem of finding an item that matches best with a user’s ...
Multimodal Educational Perceptive User Interfaces Perttu H?m?l?inen Helsinki University of Technology P.O.Box 5400, FIN-02015 HUT perttu.hamalainen@hut....
H5.2 User Interfaces EvaluationMethodology. The present study compares an adaptive simulated cellular-phone based recommender system to a non-adaptive one,...
User Interfaces for Mobi... 15页 免费 User Interfaces for Map ... 17页 免费 User Interfaces for Digi... 暂无评价 10页 免费 Parallel User ...
User Interfaces General Terms In this paper we describe the first version of the International Children’s Digital Library (ICDL). As a fiveyear research...
热门文档