Data Wrangling: Getting Started Working with Data for Visualizations
summary
Argues that any data can be visualized if it is consistently modeled and cleaned, shows that a single view realistically encodes four dimensions per view (rarely more than five to seven), and demonstrates a live project: GPS drives across eight cities in three countries visualized with layered timelines, linked tables and brushing and linking so selections highlight across views. Uses Python for wrangling and evaluates D3.js (good for single views, weak for huge datasets), praises Improvise for heavy coordinated views.
AI-written summary · about
chapters (11)
- 0:00Intro
- 0:26Rachel Shadoan
- 5:57Structuring Data Abstract Data Models
- 10:58A dimension is a variable in the data
- 15:00Dimension count: 1
- 18:41Quantitative Data
- 22:49Relational (Network) Data
- 27:31Filtering, Brushing Linking
- 36:50Improvise
- 38:23Python
- 40:12Credits
related talks
34mVisualising Multi Dimensional Data
48mData Visualization in Practice
42mInteractive Data Analysis: Visualization and Beyond
32mData Viz for Engineers: Optimizing Insight & Decision Making Through Visualization
33mData Science & Data Visualization in Python
48mVisual Storytelling w/D3-A Conceptual Intro to Visualizing Data w/JavaScript
26m"Where Numbers Meet Art: Building Data Visualizations In The Modern Web"
03h 09mEffective Data Visualization