Inform Visualization CS 7450
Popular in Course
Popular in ComputerScienence
This 0 page Class Notes was uploaded by Alayna Veum on Monday November 2, 2015. The Class Notes belongs to CS 7450 at Georgia Institute of Technology - Main Campus taught by John Stasko in Fall. Since its upload, it has received 50 views. For similar materials see /class/234085/cs-7450-georgia-institute-of-technology-main-campus in ComputerScienence at Georgia Institute of Technology - Main Campus.
Reviews for Inform Visualization
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 11/02/15
Interaction amp Dynamic Queries 2 if i 7 CS 7450 Information Visualization February 8 2005 John Stasko Case Study SDM Developed at CMU 0 Limitations of static visuals Unable to focus on different objects with context Cmuaandomaiomm3m1 OmamamdwmmdmmexmemrmeHMm data set FmHod Wommtwm Difficult to compare quantities which are not 5 atiall conti uous p y g Chuah et al UIST 95 Spring 2005 CS 7450 2 System in Action Spring 2005 cs 7450 Viewing occluded objects Interactive augmentation Spring 2005 CS 7450 4 SDM Components Objectcentered selection 0 Dynamic and flexible operations 0 SDM handles Objects constraints and feedback techniques c Context persistence c Setwide operations Spring 2005 cs 7450 Advantages Spring 2005 cs 7450 Advantages Focus and context 0 View occluded objects View different object sets in different scales Visualization is not part of the underlying data Compare the widths and heights of objects at a dista nce Spring 2005 cs 7450 Disadvantages Spring 2005 cs 7450 w Disadvantages Object manipulation causes occlusion Users cannot operate on parts of objects 0 With limited scope certain objects are occluded by the focus of objects 0 Changing the widthheight can blur the linear relationships between objects Spring 2005 CS 7450 9 Dynamic Queries 2a A more interactive query operation providing 4 Intuitive feel for the data Immediate feedback 4 Incremental reversible operations Spring 2005 CS 7450 10 DQ Sliders Potential disadvantages Operations are global in scope Controls must be fixed in advance Operations are conjunctive each filter is and ed together To see a disjunctive query you must do each operation sequentially Spring 2005 CS 7450 11 Movable Filter Magic Lens Allows interaction with more focused region of data More directmanipulationish Interaction occurs on topquot of data Fishkin amp Stone CHI 95 Spring 2005 CS 7450 12 Magic Lens I o Arbitrarilyshaped usually rectangular region with some operation that changes the user s view of the data Movable Stackable Augmented by parameters that control the display Spring 2005 cs 7450 In Action Video amp Die Len3 Spring 2005 cs 7450 Magic Lens Example J x m explain I k vinylh II Liana or oo mm X7336 i httpwwwpa rcgtlterogtltcomistp rojectsMagicLen ses Spring 2005 cs 7450 Stacking Filters 0p1 0p2 0p1 gt 0p2 0p1 0p2 Spring 2005 cs 7450 0p1 OR 0p2 0p2 AND 0p1 u Composition Manipulating stacks of lenses can be awkward Can make a compound lens by abstracting a stack of lenses into one new composition lens Spring 2005 CS 7450 17 Magnifying Lens Median Home Price 6 OO Spring 2005 CS 7450 18 Callout Lens Median Home Price Spring 2005 cs 7450 I Decatur D Marietta D Roswell I Smyrna Applying Real Values Median Home Price 160000 I Decatur I Marietta ll Roswell I Smyrna Value of some variable of data can be mapped to 00gt10 and shown visually Spring 2005 cs 7450 Other Applications What other kinds of things could you do Spring 2005 CS 7450 21 Other Applications Definition of word Details of graphical object Particular attributes of data points Spring 2005 CS 7450 22 DQ Via Magic Lens Advantages o Disadvantages Advantages H3 Liveness Flexibility Ability to specify complex queries Don t use as much real estate for controls Spring 2005 CS 7450 24 Disadvantages More complex than DQ sliders Not quite as easy to learn and use 0 More difficult to implement HW4 Update p at Any problems issues questions InfoZoom local version o How to submit Spring 2005 CS 7450 26 Final Project Rough idea of topics 0 Start forming your groups 0 Topics due Feb 17 Spring 2005 CS 7450 27 Upcoming Overview and detail Reading Chapter 7 Plaisant et al 0 HW4 discussion amp design exercise HW due in one week Spring 2005 CS 7450 28 Hierarchies and Trees 2 Spacefilling h I John Stasko Georgia Institute of Technology Hierarchies E A I I 0 Definition Data repository in which cases are related to subcases Can be thought of as imposing an ordering in which cases are parents or ancestors of other cases InfoVis 2 Last Time NodeLink Reps T Keaxns can Shannan ch12 Executlve Efflcex SpaceTree ACME TreeMmers Vice Flesidenl Near Edge PC M anzger Tech Am A Eastern DMslun Ear Mani Western Land PC Manager 5 I SuppunFunctxuns Operatinns Manager ops Ass 92 Plaisant Grosjean amp Bederson InfoVis Lamping amp Rao m A if quot1 Card Mackinlay amp Robe tson FlexTree u l Ln L1 7 E j A 1 a 1 l e m I 7 7 amp Nation 337739 397 7quot 77 39 i rgri Lonnoaes 1137quot ll L233 L3 13 L44n L53n ll Lean Figure 6 Full tree vlew of FlexTree r the structure of the tree is fully revealed Song Curran amp Sterri3tt Nodelink Shortcoming o Difficult to encode more variables of data cases nodes Shape Color Size but all quickly clash with basic nodelink structure InfoVis 4 e A SpaceFilling Representation Each item occupies an area Children are contained under parent m One example InfoVis 5 Treemap 4 Spacefilling representation developed by Shneiderman and Johnson Vis 91 0 Children are drawn inside their parent 0 Alternate horizontal and vertical slicing at each successive level 0 Use area to encode other variable of data items InfoVis Treemap Example Directories InfoVis 7 Treemap I zip Color default InfoVis Focus level 5 Up One Directorgl Topl H Bottom level shown 1 2 Higher Deeper 221 lType Rae IRandoml 1 Maximum I I hbnff a Click an amglnn mm m menmy twice m mucus File and directory Visualizer Treemap Algorithm Draw Change orientation from parent horizvert Read all files and directories at this level Make rectangle for each scaled to size Draw rectangles using appropriate size and color For each directory Make recursive ca using its rectangle as focus InfoVis 9 Nested vs Nonnested Nested TreeMap InfoV is 10 Applications Can use Treemap idea for a variety of domains Filedirectory structures Basketball statistics Software diagrams Tennis matches InfoVis Software Visualization App Eariquot 94 a SeeSys Software Metrics Visualizing System 0 Uses treemapIike visualization to present different software metrics 0 Displays Size Recent development High fixonfix rates History and growth Baker and Eick InfoVis 95 Sample View 1 Subsystems in a software system Each rectangle represents the noncomment source code in a subsystem Area means size l New code In this release InfoVis Sample View 2 Bug rates by subsystem and directory riewcode Y V 3 m quot7 Added inthis f t39 39 w Iii w Bars represent individual L directories in the subsystem InfoVis 14 Tennis Viewing Application r ar 5101 Analyze review and browse a tennis match Spaceflllingtreemaplike hierarchy representation for a competition tree Shows matchsetsgamespoints Uses lenses to show shot patterns Redgreen to encode two players Composite colors on top of each other Jin and Banks EEE CGampA 97 InfoVis 15 Visualization Makeup air InfoVis 16 Simulated Match Results Match view Bond won Set results Lens showing ball movement on individual points Game results InfoVis 3333 zmltltm mac om zm mnm 08 mp mag 32824 S oSm Hm Treemap Affordances 4 RWF Good representation of two attributes beyond nodelink color and area 0 Not as good at representing structure What happens if it s a perfectly balanced tree of items all the same size Also can get longthin aspect ratios Borders help on smaller trees but take up too much area on large deep ones InfoVis Aspect ratios InFoVis These kinds of rectangles are visually unappealing Which has bigger area 20 Variation A 5 0 Can rectangles be made more square think about it o In general a very hard problem InfoVis 21 Variation Cluster Treemap FE zdr39 SmartMoneycom Map of the Market Illustrates stock movements Compromises treemap algorithm to avoid bad aspect ratios Basic algorithm divide and conquer with some hand tweaking Takes advantage of shallow hierarchy www smartmoney commarketmap Wattenberg Image on next slide CHI 99 InfoVis 22 mmmmmammm Nasdaq3595272537ou7wo 232m on 3 Early Retirement s llanafac ln 2quot Health Care Financial Basm Ma Chm here Eommunlmm Fezlumd Adver ser E ng yi Md mummy wrvung InfoWs H EH2 gm mew ngurwes Inn s e p 53 agea mh Emma Swmv g Q regs a my WWW smartmnney Dmmarketmap TAKE A SHORT SURVEY AND W N K30 IN IOO WINS GUARANTEE D V 3 mp Ma x MapYaurPnnJaQiu 5mm Enemy Lilli KELLY dB M I I u EH2 gm ew ngumes 1m HE p quotFaack 1 v 1 a may Erasmus Mema gv av 1 Q Mawwmm mmuwmn ATI39ENTION an mm a b SMALL ELmNESs owyms L u unon hrynur bu ADWy m n summmugmm A bad day A InfoVis SmartMoney Review E 5 Tufteesque micromacro view 0 Dynamic user interface operations add to impact 0 One of best applications of InfoVis techniques that I ve seen InfoVis 26 Other Treemap Variations F 4 Ew39 o Squarified treemap Bruls Huizing van Wijk EuroGraphics 00 Alternate approach similar results Cluster Squari ed InfoVis 27 Square Algorithm Problems 4 RWF Small changes in data values can cause dramatic changes in layout 0 Order of items in a group may be important InfoVis 28 New Square Algorithms o Pivotbysize and pivotbymiddle Partition area into 4 regions Pick pivot element Rp Size Largest element Middle Middle element R1 elements earlier in list than pivot R2 elements in list before R3 and also that makes Rp have aspect ratio closest to 1 Shneiderman amp Wattenberg InfoVis 01 InfoV is Piuot bymiddle m m Piuotby size New Variation Strip treemap Strilerreemm Use strips to place items Put new rectangle into strip If it makes average aspect ratio of all rectangles in strip go down keep it there If it makes aspect ratio go up put it back and move to next strip Bederson Shneiderman amp Wattenberg ACM Trans on Graphics 02 InfoVis 30 www cs umd eduhciltreemaprhistoryj avaialgorithmsLayoutApplet html Compare results mg n Compare slice and dice squarified strip pivot techniques by aspect ratio width to height structural change metric designed to measure movements of items readability 7 WM metric based on changes Avg Reagggiivi Avg Readabiiny Avg Readahiinyu72 Avg Readahiinyui7 in direction of eye gaze as E items scanned I g grim 2i 9J1 rD um n InfoVis 3 1 Avg Aspg 7125 Avg Aspga2za Avg Asped255 HI 57Hle SIiceanddice Pivotbymiddle InfoVis Showing Structure n A 5101 Regular borderless treemap makes it challenging to discern structure of hierarchy particularly large ones Supplement Treemap view Change rectangles to other forms InfoVis 33 Variation Cushion Treemapw Van Wijk amp van de Wetering Add shading and texture InfoVis 99 to help convey structure of hierarchy InfoV is wwwwintuenlsequoiaview File Visualizer built using cushion treemap notion 39 i iHHHI iiiliiIIiIIIUWM ii Ii39ll i Wm WWWWWH Mi iil Iii W lllllllill39lll l l39quotIliliiil39 Hi ii iiillliil Demo InfoVis 35 www hivegroup comamazon html Product Sales The Hive w amp HDTV 207mm to 257mm Ammsmm mum th 12m 4213mm j mum r Ouance 1 WWW r smmd I39 ah i n r IuL cgumm l7 Puce v InfoVis 36 wwwmarumushicomappsnewsmapnewsmapcfm News Stories 2 r I EJEJ PJI mg m Mew mules m M 1 QBatk v Q a fl Seavh imam x 1quot agavesdampuwww mavumusfjfamlappslnewsmapnewsmapfm U gen 7 gt W pmm V Hamwmjzm Cm 39 l Memorial Karaml Says ka my FEW Boeing leader Jll ue39s ain mother Stres M mm m steadiness quotT 5W Apathy 5mm suiigi vm Hackers Is quot g ggsaah arequot nilLexis Taiwan Plans J r Bmmimm 112 Mllavvseim n s asxi39ifc 39i is were 355 u ggknws W 7 7 r t 3 mg m mm mm new leader W mth Mnumst a w m quot39 quotn39 quot alikan Halens w s nnwuumx mumm amdinaj Hong Kong gm bmsl H1 departs for quot 39 mmmyt ul e The Hague quot39 V39 mm mnmm r v s V I m m V I 1 Mg m 7 z n 39 3quot quot3 wrrur r u 5 mm mm WNW 53 msgum mu mum V i quot Anzasm Mulmr r y El mm mm en V m Database 5 by m r mu 1 7 am gran m M w mmch W Amancan dinner gumquot snack F mumquot my man mum uum m Mum u Wuwmwz mm mum mm ammmmu mm Thursday March 10 2005 51a Em v m m m w m rm mm m Hmqu W rm mm W W a m o mm InfoVis 37 www panopticon com Investment Portfolios Dz Elnse Window a mum suwma mum mum Panopticon suwnkls mm anvLucaLsErML Ea nunu mm mm mum cmAsuMEnw SEVEIIMES mmmuummm x m V mum swam man m m m Wm mm 3amp1 mm nquot 1 mm mm cw nr 2 O Banankun a Inlemet j g1 Dane InfoVis Another Problem FE zdr39 o What if nodes with zero value mapped to area are very important Example Stock or mutual fund portfolios Funds you don t currently hold have zero value in your portfolio but you want to see them to potentially buy them InfoVis 39 n A FundExplorer Show mutual fund portfolios including funds not currently held Area maps to your relative investment in fund 0 Want to help the user with portfolio diversification as well If I add fund X how does that overlap with my current fund holdings Csallner Handte Lehmann amp Stasko InfoVis 03 InfoVis 40 Solution F 4 Ew39 Context Treemap Treemap with small distortion Give zerovalued items all together some constant proportion of screen area Provide dynamic query capabilities to enhance exploration leading to portfolio diversification InfoVis 41 FundExplorer w e x 5 E D Exam 55 3D E Demo Voronoi Treemaps Balzer amp Deussen InfoVis 05 InfoV is 43 wwwcsumdeduhciltreemap history The World of Treemaps mi Maryland HCIL website devoted to Treemaps Workshop in 2001 there on topic InfoVis ml n E s Lat Mew 5a gammmam ew 3 a is Back meavd Retaad Ham Seav New 2 Pm Sammy 5m gt BaakmgvK A anahnnhupwawcsumdeduhc Ueemaps v wmmexatea pannersmps Contact HumanCompuur lnurac an th Unlvamlly a Maryland mam hers quota home research pubncauons academics Treemapsfor 39 quot quot quot ofquot by Een Shnexdemmn December 261998updatedhnuuy 29 mm abom hell Our keemap products xeemag z D General keemap m1 Free am msmn pxus hcensmg mfammuanfmfu package PhataMesa Zaamable image hbrgg bmwsex xee am msmn pxus hcensmg mfamauanmepackage xeemag Axgmmms and Algorithm Ammums open source Java code A Histnry of Treemnp Research m K119 University nrMmylm detemnmngwhnch um cansmamg Mg 5th arm we we a cmmks r 2x1 m but u mk m ch TR 9mg Chaasmg the right name mk probably as lung but the term keemap described the mum of bscxtauansaumalsm 199mm glrsmexdem acer lhng mp u www sum or u tummg atquot m spun 5p y mu amgn mpxy nested Lhe anges but a mare an eff m muman skategytaak only afew mums but mm TheT av 99 ca ammmmmcm hmau andledtathewxdelymedpapexGCILTRglr gjmnuyauth HQMJ A 44 Another Technique 4 RWF What if we used a radial rather than a rectangular spacefilling technique We saw nodelink trees with root in center and growing outward already 0 Make pietree with root in center and children growing outward Radial angle now corresponds to a variables rather than area InfoVis 45 Appears I 39n Americar He 39tsge Dcb39onay 3rd Ed Houghmn Mif in 1992 Radial SpaceFilling Chuah InfoVis 98 Andrews amp Heidegger InfoVis 98 InfoVis 47 SunBurst aux I sty l ca re El di recta ry I executame defau U InfoVis Treelnap Directory Visualizer Stasko Catrambone Guzdial amp McDonald JHCS OO 48 SunBurst n A 5101 Root directory at center each successive eve drawn farther out from center Sweep angle of item corresponds to size 0 Color maps to file type or age 0 Interactive controls for moving deeper in hierarchy changing the root etc 0 Doubleclick on directory makes it new root InfoVis 49 Compared SunBurst to Treemap borderless on a variety of file browsing tasks SunBurst performed as well or better in task accuracy and time Learning effect Performance improved with Treemap on second session Strong subjective preference 519 for SunBurst Participants cited more explicit depiction of structure as an important reason More to come on evaluation InfoVis 50 Empirical Study SunBurst Negative if at o In large hierarchies files at the periphery are usually tiny and very difficult to distinguish E examples 7 InfoVis 5 1 F 4 Fix Objectives IVlake small slices 0 Avoid use of multiple blgger windows or lots of 0 Maintain full circular scrollbars spacefilling idea 0 Allow detailed examination of small files within context of entire hierarchy 0 Don t alter ratios of sizes 0 Provide an aesthetically pleasing interface in which it is easy to track changes in focus InfoVis 52 FE Ad 0 Three visualizationnavigation techniques developed to help remedy the shortcoming Angular detail Detail outside Detail inside 3 Solutions Stasko amp Zhang InfoVis 00 InfoVis 53 Angular Detail H 7 i 7 aquot 0 Most natural 0 Least spaceef cient 0 Most configurable by user InfoVi s 54 Detail Outside a a o Exhibits nondistorted miniature of overview 0 Somewhat visually disconcerting 0 Focus is quite enlarged large circumference and 360 0 Relatively space ef cient InfoVis 55 Detail Inside f a e an 0 Perhaps least intuitive and most distorting 0 Items in overview are more distinct larger circumference 0 Interior 360 for focus is often sufficient InfoVis See in Action iii Video InfoVis 00 In fDVis 57 Key Components n A 5101 Two ways to increase area for focus region larger sweep angle and longer circumference Smooth transitions between overview and focus allow viewer to track changes 0 Always display overview Allow focus selections from anywhere normal display focus or overview regions InfoVis 58 SunBurst o Demonstration of system Javabased version Developed by Neel Parekh Reads XML specification of hierarchy Incorporates some of the animated focus operations InfoVis 59 Potential Followon Work F 4 Rw39 Multiple foci o Varying radii for different levels in hierarchy Use quickkeys to walk through neighboring files Smarter update when choosing new focus region from existing focus 0 Fourth method expand angle of focus in place by compressing all others InfoVis 60 InterRing Provides many of those followon capabilities and new operations Yang Ward amp Rudensteiner InfoVis 02 InfoVis 61 More Alternatives Combine spacefilling hierarchy presentations really nesting with zooming Children drawn inside of parent but not totally encompassing InfoVis 62 wwwgroxis com W em 1 E R m w E umemmswa Yava 5 2 grim he p A E x Demo Mame We Deskkav Favantes Mv Dacuments 5m Menu UsevData P uquL S hate was Fue w BL L7 L123gtL1L7 L L me u Desmptmn zoniams InfoVis Zoomology j jm CS 7450 Spring 03 project InfoVis 03 Contest Winner Best Student entry InfoVis 64 Alternate View Video InfDVis 65 Hybrid Approaches 0 Mix nodelink and spacefilling InfoVis 66 CHEOPS 7777 7 Bengraphy Histnry History Prehistory Evo uti an Saw in previous lecture InfoVis Beaudoin Parent Vroomen Vis 96 67 EnCon 0 Goal Vis for large and hierarchical data sets 0 3 key issues for the design of effective interactive visualizations Efficient Space utilization specially in large data sets Navigation mechanisms visual is useless without good navigation Minimization of human cognitive process see both relationships and details Nguyen amp Huang Information Visualization 05 InfoVis 68 i 1 s i I quot i quot V I x f L i s i R I I I I I I I IIIIII Connection Approach Enclosure Approach IZlImmediate perception IZlCapable to display large of relationships amounts of information Not efficient in terms In limited Space of utilizing space Don t show relationships of information structure InfoVis 69 Partitioning and Nodes I R 7 f Fast algorithm to calculate the geometrical layout partitioning Nodelink diagram that 9 DJ represents hierarchical 0 0 structure I O O O O InfoVis 70 Navigation Techniques 0 Access both Contextual ancl Local Information 0 Techniques Focus Context distortion or semantic zoom detailfocus views Zooming Filtering Incremental Exploration InfoVis 71 Summary 4 RWF Nodelink diagrams or spacefilling techniques 0 It depends on the properties of the data Nodelink typically better at exposing structure of information structure Spacefilling good for focusing on one or two additional variables of cases InfoVis 72 Interaction amp Dynamic Queries 2 I CS 7450 Information Visualization February 6 2003 John Stasko Dynamic Queries A more interactive query operation providing Intuitive feel for the data Immediate feedback Incremental reversible operations Spring m a mu 2 DQ Sliders Potential disadvantages Operations are global in scope Controls must be fixed in advance Operations are conjunctive each filter is and ed together To see a disjunctive query you must do each operation sequentially Spring m a mu 2 Movable Filter Magic Lens H I Allows interaction with more focused region of ata More directmanipulationish Interaction occurs on topquot of data Fishkin amp Smne I 95 Spring m a mu Magic Lens Arbitrarilyshaped usually rectangular region with some operation that changes the usel s view of the data Movable r Stackable r Augmented by parameters that control the display Spring m a mu In Action Video quot 333 Die Larisa Spring m a mu 5 Magic Lens Example Stacking F39 ers l l mun ll mum compo n g aWKWard a 5m Bf 525 mu Una new Dmusmun ms Magnifying Lens Median Home Price Swing m a mu m Callout Lens Media n Home Price Swing m a mu n Applying Real Values Price mu Value of some variable of data can be mapped to 00gt10 and shown visual y Swing m a mu m Other Applications What other kinds of things could you do Spring m a mu u Other Applications Definition of word Details of graphical object articular attributes of data points Spring m a mu m DQ Via Magic Lens Advantages Disadvantages Spring m a mu 5 Advantages Liveness Flexibility Ability to specify complex queries Don t use as much real estate for controls Spring m a mu ls Disadvantages More complex than DQ sliders Not quite as easy to learn and use More difficult to implement Spring m a mu n A1 Thoughts Informal discussion of systems Spotfire eeIt Eureka Spring m a mu xx Interaction amp Dynamic Queries 2 if i 7 CS 7450 Information Visualization February 6 2003 John Stasko Dynamic Queries a o A more interactive query operation providing 4 Intuitive feel for the data Immediate feedback 4 Incremental reversible operations Spring 2004 CS 7450 2 DQ Sliders Potential disadvantages Operations are global in scope Controls must be fixed in advance Operations are conjunctive each filter is and ed together To see a disjunctive query you must do each operation sequentially Spring 2004 CS 7450 3 Movable Filter Magic Lens Allows interaction with more focused region of data More directmanipulationish Interaction occurs on topquot of data Fishkin amp Stone CHI 95 Spring 2004 CS 7450 4 Magic Lens I o Arbitrarilyshaped usually rectangular region with some operation that changes the user s view of the data Movable Stackable Augmented by parameters that control the display Spring 2004 cs 7450 In Action Video amp Die Len3 Spring 2004 cs 7450 Magic Lens Example J x m explain I k vinylh II Liana or oo mm X7336 i httpwwwpa rcgtlterogtltcomistp rojectsMagicLen ses Spring 2004 CS 7450 7 Stacking Filters Opl 0p2 Opl 0p2 Opl 0p2 Spring 2004 CS 7450 8 u Composition Manipulating stacks of lenses can be awkward Can make a compound lens by abstracting a stack of lenses into one new composition lens Spring 2004 CS 7450 9 Magnifying Lens Median Home Price 6 OO Spring 2004 CS 7450 10 Callout Lens Median Home Price Spring 2004 cs 7450 I Decatur D Marietta D Roswell I Smyrna Applying Real Values Median Home Price 160000 I Decatur I Marietta ll Roswell I Smyrna Value of some variable of data can be mapped to 00gt10 and shown visually Spring 2004 cs 7450 Other Applications What other kinds of things could you do Spring 2004 CS 7450 13 Other Applications Definition of word Details of graphical object Particular attributes of data points Spring 2004 CS 7450 14 DQ Via Magic Lens Advantages o Disadvantages Advantages H3 Liveness Flexibility Ability to specify complex queries Don t use as much real estate for controls Spring 2004 CS 7450 16 Disadvantages More complex than DQ sliders Not quite as easy to learn and use 0 More difficult to implement HW3 Update p at c Any problems issues questions How to submit Spring 2004 CS 7450 18 Final Project Rough idea of topics 0 Start forming your groups 0 Topics due Feb 19 Upcoming g HW3 discussion amp design problem Time Series Data Overview and detail Spring 2004 CS 7450 20 Hierarchies and Trees 2 A sin 5 74 r Infurmatmn szJahzaunn F urua39y 27 Jnhn En ku Hierarch39 s Dz nmnn vevasxtawm Wm cases ave re ahed m are Suhca mauqht D g WW5an an mde m Wm 25 ave Davents m ancesmvs afathev E nk Reps Last Time39 Node Nodelink Shortcoming Difficult to encode more variables of data cases nodes but all quickly clash with basic nodelink structure Spring m a mu SpaceFilling Representation Each item occupies an area Children are contained under parent m One example Swing m a mu 5 Treemap Space filling representation developed by Shneiderman and Johnson Vis 91 Children are drawn inside their parent Alternate horizontal and vertical slicing at each successive eve Use area to encode other variable of data items Spring m a mu Treemap 5 Emma mm Dema Ne 2nd ammmhm Treemap Algorithm M lt mm mum fwm pm mamma m 1 Me And Mam 2 m m We mm mm mm w m Duwvezum e mg wwpmm 2nd za av Fm w amw We mm lt2 mm m mm 1 m gt Nested Vs Nonnested swam d amt m wae mamas D splays m deveavmm Hugh xranr x Vales 5m and qmwth Sample View 1 Sample V39 w 2 ag Mmmmmmmww a math mpresmemn fur a mmpemun tree Uses Enss m shm m pattams Redgreen m mqu WU players w Treemap Affordances Good representation of two attributes beyond nodelink color and area Not as good at representing structure What happens if it s a perfectly balanced tree of items all the same size Also can get longthin aspect ratios 7 Borders help on smaller trees but take up too much area on large deep ones Spring m a mu m Aspect ratios v hese kinds of rectangles are visually unappealing Which has bigger area Spring m a mu zn Variation i ia Can rectangles be made more square think about it In general a very hard problem Spring m a mu 2 Clustef 39reemap Hustvahes 5m mavavv m tampmmeyneemap a qanthm m avmd had aspect vahas same hand Meamq m gammy commmmp SmanMoney Rev w Dynamw user nterface Uperatmns add an act v One u nesi apphtatmns man cemmques that We seen Other Treemap Variations A m Equan ed mmap mg Hmzwq van ka News m h Square Algorithm Problems g a reman changes n aDut mpc ant g w amp usemap compare results 7 gt mg a 2nd due om 2mg n39m by mad Show39ng Structure W haHengmg m mscem 5mm m hwaamhy panmlar y 3792 was Wequot mt Yveemav Mew Varia 39on on 39on Treemap a Sequoiav w m mxm g m m quzhxd mm um um lveemiv The World of Treemaps g m Maw12nd H01 Mme mm In new wmmp m znm there an W Another Technique redawgma watermhng technque r We saw naderhnk K1225 W mu m mm and yummy autwavd may Malt2 pErh39EE W mat n Ema am Miran gmwwng uutward vathev than 5125 hush Andvews amp Hexdeqqm mm ya SunBurst SunBurst in we drawn fanha cut rum Ema cmur mys m mg type m 592 a w hweramhy mangmg the mm at Dmbbrdtk an mam malt25 n new mat SunBurst Demnnsh39atmn Bf ssiem munmmmele an am Sun Emp cal Study quot an a may mg trnwsng asks mum am He Yveavvav an 52mm 5255er Pamdvmts mad Mme ExvhntdEDxctmn af 5mm as an vatant veasan SunBurst Negative v 2 H H n wH H are waxy my am vay mm m msungmsH Object es Mekemu hwqqev Mememmumm ey speeeaume me n a use afmu l v e wmdaws m m D snaHhas medean Dwdeta ed aestheticd eaan exammahanafsma N25 Wm mm Dr nhevface m men n 5 enhve meyem e Dania tevvatmsaf easY m hack themes m fwms 3 Solut ns se devebped m he p remedy the shunmrmng Deta meae AngularDeta CI CI 9 16157 IQN Detail Du 5mm m mm I mmmm aFmuuew CI 539 1 39i3 v a 5 me Detall Inslde mm km NW 2nd m dmamnq draw m amewm mm am we mummna me m m u when umuenl Key Components Two ways to increase area for focus region larger sweep angle and longer circumference Smooth transitions between overview and focus allow viewer to track changes Always display overview Allow focus selections from anywhere normal display focus or overview regions Spring m a mu u Potential Followon Work a Multiple foci Varying radii for different levels in hierarch Use quickkeys to walk through neighboring files Smarter update when choosing new focus region from existing focus Fourth method expand angle of focus in place by compressing all others Spring m a mu 5n Hybrid Approaches Mix node link and space filling Spring m a mu 5 CHEOPS r I CHEOPS A Compact Explorer For Complex Hierarchies CRIM39s Hierarchical Engine for OPen Search A A Beaudoin Parent Vroomen CHECIF39LS Vis 96 sum 2cm 5 745m 52 What CHEOPS Is Compressed visualization of hierarchical data using triangle tessellation Most or all of the hierarchy can be displayed at once v Since no DegreeofInterest DOI function required no major recalculation required when focus changes sum 2cm 5 745m 53 Triangle Tessellation u Overlaptile the A triangles A A The visual object 5 is overloaded with the logical nodes E and F Insert overlapping triangles between logical nodes sum 2cm 5 745m 54 18 O EOPS reuses v sua oompmems Waugh a mmate brandw de owent Growth reduced to mapqmaanc AH Davmtnmz mDth se enad as wau A a 5mg 1 a mquot 2mm Laughing naequot m m mum mm m Muwmm Nguyeme m a w m Mm m m mm mam b mmmiquot m 39wiwmm39mmm Summary Nodelink diagrams or spacefilling techniques It depends on the properties of the data Nodelink typically better at exposing structure of information structure Spacefilling good for focusing on one or two additional variables of cas Spring m a mu HW4 Drawa graph Swing m a mu HW 3 Jazz amp HiNote Will discuss after break Spring m a mu 20
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'