Lectures 3&4: from categorical and ordered Express Separate - PowerPoint PPT Presentation

How to handle complexity: 1 previous strategy + 3 more Facet How? Encode Manipulate Facet Encode Manipulate Facet Reduce Juxtapose Manipulate Facet Reduce Map Derive Arrange Change Juxtapose Filter Lectures 3&4: 
 from categorical and ordered Express Separate attributes Change Juxtapose Filter Color Facet into Multiple Views Hue Saturation Luminance Select Partition Aggregate Order Align Partition Size, Angle, Curvature, ... • derive new data to Use Select Partition Aggregate Tamara Munzner Navigate Superimpose Embed show within view Shape Department of Computer Science • change view over time University of British Columbia Superimpose Motion Direction, Rate, Frequency, ... • facet across multiple Navigate Superimpose Embed DSCI 532: Data Visualization 1I views Lectures 3&4: 27 & 29 March 2017 • reduce items/attributes https://github.ubc.ca/ubc-mds-2016/DSCI_532_viz-2_students within single view 2 3 4 Idiom: Linked highlighting System: EDV System: Pathfinder Juxtapose and coordinate views Linked views Complex linked multiform views • see how regions contiguous in one view are • unidirectional vs Share Encoding: Same/Di fg erent • linked views distributed within another bidirectional linking Linked Highlighting • simultaneously visible –powerful and pervasive interaction idiom multiple views • encoding: different • linked together such that actions in one view –multiform Share Data: All/Subset/None affect the others • rationale: single monolithic view has strong limits on number of attributes that can be shown simultaneously • data: all shared Share Navigation [Visual Exploration of Large Structured Datasets. Wills. Proc. New Techniques and Trends in Statistics (NTTS), pp. 237–246. IOS Press, 1995.] https://www.youtube.com/watch?v=aZF7AC8aNXo http://www.ralphstraumann.ch/projects/swiss-population-cartogram/ http://peterbeshai.com/linked-highlighting-react-d3-reflux/ 5 6 7 8 System: Google Maps System: StratomeX Idiom: bird’s-eye maps Overview-detail Shiny example Idiom: Parallel sets • encoding: same • data: subset shared • navigation: shared –bidirectional linking • differences –viewpoint –(size) • overview-detail [A Review of Overview+Detail, Zooming, and Focus+Context Interfaces. Cockburn, Karlson, and Bederson. ACM Computing Surveys 41:1 (2008), https://www.jasondavies.com/parallel-sets/ 1–31.] https://gallery.shinyapps.io/TSupplyDemand/ https://www.youtube.com/watch?v=UcKDbGqHsdE https://eagereyes.org/parallel-sets 9 10 11 12 System: Mondrian System: MizBee System: Cerebral Idiom: Mosaic plots Overview-detail Shiny example Idiom: Small multiples • multiscale: three viewing levels • APGI genome browser • encoding: same –tooling: processing (modern version: p5js.org) –tooling: R/Shiny • data: none shared –interactivity –different attributes • tooltip detail on demand for node colors on hover –(same network • expand/contract layout) chromosomes • navigation: shared • expand/contract control panes http://www.theusrus.de/blog/understanding-mosaic-plots/ http://www.theusrus.de/Mondrian/ https://www.youtube.com/watch?v=76HhG1FQngI&t=2s http://www.theusrus.de/blog/making-movies/ [Cerebral: Visualizing Multiple Experimental Conditions on a Graph with Biological Context. Barsky, Munzner, Gardy, and Kincaid. IEEE Trans. https://www.youtube.com/watch?v=86p7brwuz2g https://gallery.shinyapps.io/genome_browser/ Visualization and Computer Graphics (Proc. InfoVis 2008) 14:6 (2008), 1253–1260.] 13 14 15 16

Coordinate views: Design choice interaction Why not animation? System: Improvise Video: Visual Analysis of Historical Hotel Visitation Patterns • disparate frames and • investigate power of multiple views regions: comparison All Subset None difficult –pushing limits on view count, Overview/ –vs contiguous frames Same interaction Redundant Detail –vs small region complexity Small Multiples –how many is ok? –vs coherent motion of group Multiform, • open research No Linkage Overview/ question Multiform Detail • safe special case –reorderable lists • why juxtapose views? • easy lookup –animated transitions • useful when linked to –benefits: eyes vs memory other encodings • lower cognitive load to move eyes between 2 views than remembering previous state with https://www.youtube.com/watch?v=Tzsv6wkZoiQ single changing view [Building Highly-Coordinated Visualizations In Improvise. Weaver. Proc. IEEE Symp. Information http://www.cs.ou.edu/~weaver/improvise/examples/hotels/ –costs: display area, 2 views side by side each have only half the area of one view Visualization (InfoVis), pp. 159–166, 2004.] 17 18 19 20 Idiom: Trellis plots Idiom: Trellis plots Partition into views Partitioning: List alignment • single bar chart with grouped bars • small-multiple bar charts • matrix alignment for small multiple plots • main effects ordering • how to divide data between views Partition into Side-by-Side Views –split by state into regions –split by age into regions –split into regions by attributes –same issues as alignment for marks within plot! –order small-multiples plots based on derived • complex glyph within each region showing all • one chart per region data to see trends –encodes association between items ages • partition by –compare: easy within age, harder using spatial proximity –compare: easy within state, hard across ages –order plots by median values across states –year for columns –order of splits has major implications –shared vertical axis within each plot ordered by 11.0 11 65 Years and Over –site for rows (alphabetical) 45 to 64 Years 5 for what patterns are visible median values within varieties 10.0 25 to 44 Years 0 18 to 24 Years • within pane 11 9.0 14 to 17 Years • no strict dividing line 5 5 to 13 Years 0 8.0 Under 5 Years –variety for vertical axis 11 –view: big/detailed 5 7.0 0 –yield for vertical position 6.0 11 • contiguous region in which visually 5 5.0 encoded data is shown on the display 0 11 4.0 5 –glyph: small/iconic 0 3.0 11 • object with internal structure that arises 5 2.0 0 from multiple marks 1.0 11 5 0.0 0 21 22 23 24 CA TK NY FL IL PA CA TK NY FL IL PA Partitioning: Recursive subdivision System: HIVE Partitioning: Recursive subdivision System: HIVE Partitioning: Recursive subdivision System: HIVE Partitioning: Recursive subdivision System: HIVE • split by neighborhood • switch order of splits • different encoding for • size regions by sale second-level regions counts • then by type –type then neighborhood –choropleth maps –not uniformly • switch color • then time • result: treemap –by price variation –years as rows –months as columns • color by price • type patterns –within specific type, which neighborhoods • neighborhood patterns inconsistent –where it’s expensive –where you pay much more for detached type [Configuring Hierarchical Layouts to Address Research Questions. Slingsby, Dykes, and Wood. IEEE Transactions on Visualization and Computer Graphics [Configuring Hierarchical Layouts to Address Research Questions. Slingsby, Dykes, and Wood. IEEE Transactions on Visualization and Computer Graphics [Configuring Hierarchical Layouts to Address Research Questions. Slingsby, Dykes, and Wood. IEEE Transactions on Visualization and Computer Graphics [Configuring Hierarchical Layouts to Address Research Questions. Slingsby, Dykes, and Wood. IEEE Transactions on Visualization and Computer Graphics 25 26 27 28 (Proc. InfoVis 2009) 15:6 (2009), 977–984.] (Proc. InfoVis 2009) 15:6 (2009), 977–984.] (Proc. InfoVis 2009) 15:6 (2009), 977–984.] (Proc. InfoVis 2009) 15:6 (2009), 977–984.] Superimpose layers Static visual layering Idiom: Trellis plots Superimposing limits CPU utilization over time 100 • layer : set of objects spread out over region • foreground layer: roads • superimpose within same frame • few layers, but many lines 80 60 –each set is visually distinguishable group –hue, size distinguishing main from minor –color code by year –up to a few dozen 40 20 –extent: whole view Superimpose Layers –high luminance contrast from background –but not hundreds 0 05:00 05:30 06:00 06:30 07:00 07:30 08:00 • background layer: regions • superimpose vs juxtapose: empirical study • design choices 100 80 –how many layers, how to distinguish? –desaturated colors for water, parks, land –superimposed for local, multiple for global 60 areas • encode with different, nonoverlapping channels –tasks 40 20 • user can selectively focus attention • two layers achieveable, three with careful design • local: maximum, global: slope, discrimination 0 05:00 05:30 06:00 06:30 07:00 07:30 08:00 –small static set, or dynamic from many possible? –same screen space for all multiples vs single • “get it right in black and white” 100 superimposed 80 –check luminance contrast with greyscale 60 view 40 20 [Graphical Perception of Multiple Time Series. [Get it right in black and white. Stone. 2010. 
 0 Javed, McDonnel, and Elmqvist. IEEE Transactions 05:00 05:30 06:00 06:30 07:00 07:30 08:00 http://www.stonesc.com/wordpress/2010/03/get-it-right-in-black-and-white] on Visualization and Computer Graphics (Proc. IEEE InfoVis 2010) 16:6 (2010), 927–934.] 29 30 31 32