Whirlwind tour of data visualization

abstract

In this talk, I will describe data visualization from three perspectives: historical, theoretical, and software implementation. I will review the history of data visualization, starting with Enlightenment practitioners and concluding with today’s data scientists. I will review some theories of data visualization, emphasizing Jacques Bertin, Leland Wilkinson, and Hadley Wickham. I will review the software history of data visualization, emphasizing the software used by data scientists today. Finally, I will mention the main venues for data visualization today in popular media and scientific literature. After the talk, I will ask attendees to create a quick data visualization using Tableau.

But first …

How should we talk about viz?

  • catalog?
  • definitions?
  • history?
  • theory?
  • implementations?

Catalog example: cartogram

Cartogram of World Population, 2000

Another: a chord diagram of programming languages

Chord diagram of programming languages used in CRAN

Coiled bar chart of CO2 emissions

Bad visualizations in the catalog

Meaningless visualization about baby boomers

Definitions

Is this a data visualization?

Woman wearing a dress that resembles a map of Africa

The world’s tallest tree and two humans for scale

The world’s tallest tree and two humans

Venn diagrams of data viz 1

Venn diagram of info viz

Venn diagrams of data viz 2

Venn diagram of info design

Venn diagrams of data viz 3

Venn diagram of data science components

History

Pre-history:

Wait for Michael Friendly’s and Denis York’s book on the subject.

Enlightenment

Florence Nightingale

Coxcomb plot of causes of death

William Playfair

time series line chart

W.E.B. DuBois

Unique Population Graphic

Context of preceding viz: photo of WEB DuBois exhibit, including diagrams, photos, and books

Visualization in the 1950s: SAGE

SAGE consoles

Exploratory Data Analysis 1970s and 80s

  • John W. Tukey
  • William Cleveland
  • Edward Tufte

Anscombe’s Quartet

Four data sets

Anscombe’s Quartet Visualized

Four scatterplots

Cover of Tukey’s 1977 book, Exploratory Data Analysis

Cover of Tufte’s 1983 book, Visual Display of Quantitative Information

Dust Jacket of Cleveland’s 1985 book, Elements of Graphing Data

Information Visualization 1990s and early 2000s

  • and its relationship to scientific visualization

Venn diagram of infoviz

Network Science early 2000s

A famous visualization of political blogs

Visual Analytics, 2005—

Book cover of Illuminating the Path (2005)

Theory

Cover of Jacques Bertin’s book, Semiology of Graphics

Cover of Leland Wilkinson’s book, Grammar of Graphics

Cover of Hadley WIckham’s book, ggplot2

Others

  • Martin Wattenberg and Fernanda Viegas
  • Jeffrey Heer

Software

Mark Lombardi’s handdrawn visualizations have been exhibited in art museums

One of many attempts to automate Lombardi’s visualizations

R running in the RStudio IDE

Python running in the RStudio IDE

Tableau

Others

  • Tableau imitators, such as …
  • Oracle OBIEE
  • Microsoft PowerBI
  • SAS Visual Analytics

END

COLOPHON

This slideshow was produced with reveal.js using the moon theme.

pandoc converted the source markdown to html.

The source is at the same URL as the presentation. Just say index.md instead of index.html

URL: https://mickmcquaid.com/whirlwind

email: mickmcquaid@gmail.com