📊

Data Visualization

41 cards · 🇺🇸 English

Chart types, plotting libraries, and data visualization best practices.

Study all 41 cards for free

Flashcards, quizzes, pixel-art games and a daily review that remembers for you.

Study this deck free

Try the first 10 cards

1 / 10
Question

Data visualization

Knew itLearning

Tap the card for the answer · swipe right if you knew it, left if not

Already have an account? Sign in

Cards

  1. 1Data visualization
    The graphical representation of information and data using visual elements like charts, graphs, and maps to make complex data more accessible and understandable
  2. 2Data-ink ratio
    Principle by Edward Tufte: maximize the ratio of data to ink. Remove chart junk and unnecessary elements. Every bit of ink should represent data. Strive for simplicity
  3. 3Gestalt principles in visualization
    Proximity: grouped items appear related Similarity: similar items appear related Continuity: eye follows continuous lines Closure: mind fills gaps Connection: connected items appear related
  4. 4Preattentive processing
    Visual properties that the brain processes automatically and rapidly (<200ms): color, size, position, orientation, shape. Use these to highlight important data without requiring conscious thought
  5. 5Main color schemes for visualization
    Sequential: light to dark for ordered data Diverging: two hues from middle for data with critical midpoint Categorical/Qualitative: distinct hues for unrelated categories Choose based on data type!
  6. 6Colorblind-friendly design
    Avoid red-green combinations (most common colorblindness). Use ColorBrewer palettes, add patterns/shapes, test with colorblind simulators. ~8% of males affected
  7. 7Chart junk
    Unnecessary decorative elements that don't represent data: 3D effects, excessive gridlines, decorative images, gradients. Coined by Edward Tufte. Remove to improve clarity
  8. 8Lie factor
    Ratio of effect shown in graphic to effect in data. Should be 1.0. Distortions include: starting y-axis at non-zero, inconsistent scales, 3D perspective making areas misleading
  9. 9Violin plot
    Combination of box plot and density plot. Shows distribution shape (like histogram rotated 90°) plus quartile summary. Better than box plot for showing multimodal distributions
  10. 10Treemap
    Hierarchical data shown as nested rectangles. Area represents quantity. Good for showing part-to-whole relationships with hierarchy. Color can encode additional dimension
  11. 11Sankey diagram
    Flow diagram where width of arrows represents flow quantity. Shows transfers between categories. Good for energy flows, budget allocation, user journeys
  12. 12Parallel coordinates plot
    Multivariate data visualization with vertical axes for each variable. Lines connect values across axes. Good for high-dimensional data, pattern discovery, and outlier detection
  13. 13Small multiples (faceting)
    Series of similar charts with same scale showing different subsets of data. Allows easy comparison across categories. Also called trellis plots or panel charts
  14. 14explanatory vs exploratory visualization
    Exploratory: for yourself, discovering patterns, trying many charts Explanatory: for audience, communicating specific insights, polished, focused on key message
  15. 15Grammar of graphics
    Framework for describing visualizations using layers: data, aesthetics (mappings), geometries, scales, coordinates, facets. Implemented in ggplot2 (R) and Plotly/Altair (Python)
  16. 16Main Python libraries for visualization
    Matplotlib: foundational, low-level, flexible Seaborn: statistical, built on matplotlib Plotly: interactive, web-based Altair: declarative, grammar of graphics Bokeh: interactive dashboards
  17. 17Main JavaScript libraries for visualization
    D3.js: powerful, low-level, data-driven Chart.js: simple, canvas-based Highcharts: commercial, feature-rich Plotly.js: scientific, interactive Vega/Vega-Lite: declarative grammar
  18. 18D3.js
    Data-Driven Documents: JavaScript library for manipulating DOM based on data. Binds data to elements, applies transformations. Low-level but extremely flexible. Industry standard for custom web visualizations
  19. 19Tableau
    Commercial BI tool for interactive data visualization and dashboards. Drag-and-drop interface, no coding required. Connects to many data sources. Industry standard for business analytics
  20. 20SVG vs Canvas
    SVG: vector, DOM elements, good for <10k objects, responsive, accessible Canvas: raster, pixel manipulation, better performance for many objects, no DOM, requires redraw
  21. 21Brushing and linking
    Interactive technique where selecting data in one chart highlights corresponding data in other charts. Enables exploration of multivariate relationships
  22. 22Choropleth map
    Map with regions colored/shaded according to data values. Common for showing geographic patterns. Warning: can be misleading with different sized regions (population density better than totals)
  23. 23Word cloud
    Text visualization where word size represents frequency or importance. Often criticized as chart junk - bar charts usually better. Use sparingly, mainly for quick overviews
  24. 24Overplotting and how to handle it
    When data points overlap, hiding density. Solutions: transparency/alpha, jittering, hexbin plots, 2D density plots, sampling, aggregation
  25. 25Aspect ratio and why does it matter
    Width-to-height ratio of chart. Affects perception of slopes in line charts. Banking to 45° (Cleveland): average slope should be ~45° for optimal perception. Consider using golden ratio (1.618)
  26. 26Waterfall chart
    Shows cumulative effect of sequential positive and negative values. Good for financial analysis (profit bridges), showing how value builds up or breaks down over categories
  27. 27Gantt chart
    Bar chart showing project schedule over time. Horizontal bars represent tasks with start/end dates. Shows dependencies and overlaps. Standard in project management
  28. 28Funnel chart
    Shows stages in a process with decreasing values. Each stage represented by progressively smaller bars/areas. Common in sales/marketing for conversion funnels
  29. 29Bullet graph
    Compact bar chart showing performance against target. Designed by Stephen Few as alternative to gauges. Shows actual, target, and qualitative ranges (poor/good/excellent) in small space
  30. 30Simpson's paradox in visualization
    Trend appears in groups but reverses when groups are combined (or vice versa). Always check for confounding variables. Visualize subgroups separately to reveal hidden patterns

And 11 more cards. Study the whole deck free

Articles about this deck

More Programming decks

See all
Study all 41 cards free