Skip to content
PostData Analysis / Data Visualization

DV-01-Introduction

2025-12-30
Back to Blog
  • 什么是数据可视化?
  • 为什么要做数据可视化?
  • 数据可视化的理论模型与语言
  • 数据可视化的发展历史与分支

DV-01-Introduction

  • What is data visualization?
    • The use of computer-supported, interactive, visual representations of data to amplify cognition.
  • Why do we need data visualization?
    • Data visualization tools are key to ‘big data’ analysis
    • One picture is worth ten thousand words
    • Statistics may mislead the users
    • Optical illusion: Use color, light and patterns to create images that can be deceptive or misleading to our brains.
    • Reading a picture is faster to grasp the information than reading the text for humans. So we need to study the human perception system and learn how to design a good visualization.

DV Theory Model

DIKW 模型(概念 + 理解)

Data → Information → Knowledge → Wisdom(Ackoff, 1989)

你需要会:

  • 每一层是什么意思
  • 数据可视化主要解决哪一层(👉 Information / Knowledge)

数据可视化的模型(重点中的重点)

van Wijk 的 Visualization Model(2005)

要点:

  • 对象(Squares):D(data), S(specification), I(image), K(knowledge)
  • 行为(Circles):V(visualize), P(perceive), E(explore)

👉 考法:

  • 图中某个字母代表什么
  • “perceive”和“explore”有什么不同

Information Visualization Reference Model(Card et al.)

流程要能说清:

shell
Raw Data Data Tables Visual Structures Views Human

中间有:

  • transformations
  • visual mapping
  • interaction

Grammar of Graphics

核心作者与时间线(会考配对):

  • Wilkinson(1999):Grammar of Graphics
  • Wickham(2005):ggplot2
  • Wickham(2010):Layered Grammar of Graphics

核心元素:

  • Data
  • Geometry
  • Scale
  • Aesthetics
  • Coordinates
  • Facets

三、历史部分怎么复习最省力(不死背)

记住 “人 + 技术 + 目的” 三件事即可。

关键人物(必认)

  • William Playfair:折线图、柱状图、饼图
  • Florence Nightingale:玫瑰图(说服)
  • John Snow:霍乱地图(分析)
  • Charles Minard:拿破仑远征(多维信息)
  • John Tukey:EDA(探索性数据分析)
  • Jacques Bertin:视觉变量理论

👉 考法:

“Who used visualization to persuade?” “Which visualization supports hypothesis generation?”


四、三大分支(几乎必考)

分支数据类型关键词
SciVis空间数据volume, flow
InfoVis抽象数据network, hierarchy
Visual Analytics分析推理human + machine

一句话记忆法:

SciVis 看空间,InfoVis 看结构,VA 看推理