Tools for Visualization

By Sarah Pittroff

Presentation slides for this topic

https://studiengang-digitale-methodik.pages.gitlab.rlp.net/modul-5/5c/slides/visualisierung/2022/

Canon(s) of Digital Cultures – Data Visualization

01 What is Data Visualization?

Possible answers:

What is Data Visualization in the Humanities?

Timelines:

or maps:

and oftentimes networks:

What is Data in the Humanities?

  • (Research) Data
    • digitized object data
    • texts
    • images
    • sound
    • moving pictures
  • Meta Data
    • creation date
    • digitazion date
    • author
    • locality
    • iconography

From variables to visuals: What kind of data can be visualized?

  • Structured data as

table, database, network, geo-spatial, time series, 3D model

  • Unstructured data as

text, image, video, sound -> must be structured to be visualized :D

Structured data (variables) goes to+ visual/graphical form (visuals)

What is data visualization good for?

  • Tool for handling amounts of data or information that are either so large or so complex that the human mind cannot oversee them without external tools
  • Generate knowledge through explorative analysis from large (unstructured) data corpora
  • Finding errors in the data set
  • Showing an argument

Reasons to Visualize Scientific Data

Visualization of data in different functionalities for the research process:

Explorative analysis

I am interested in the data set in general and try to pose undirected questions to the data by different designs of visualization (–> targeted array with different parameters that are changed according to a certain pattern).

Confirmatory analysis

I am specifically pursuing an assumption and would like to work out a significance (accumulation) or correlation (context). Perhaps I have worked out a clue by explorative data analysis, which I can now track down with a modified data set in a confirmatory analysis.

Explanatory Visualization

Here I present data to a public. The focus is on a very specific aspect, which I would like to present visually within the scientific argumentation. As the term “explanation” already says, an argument, a relationship is explained, but the visualization also has a proving character.

Explorative and confirmatory analysis foster gain in knowledge, explanatory analysis explains this knowledge.

02 Methods

Visualization of information as in different points of view:

  • Macroscopic level shows overarching structures (Distant Reading)
  • Microscopic level enables detailed views on data sets

Visualization of information is interconnected in the resarch process

  1. Data Sampling
  2. Analysis based on algorithm
  3. Visualization of data
  4. Hermeneutic interpretation of visualization

These four steps are interdependent phases. They have to be thought of as biderectional dependencies and can’t be imagined as seperated actions. For the sake of a good visualizaton we have to take care of two premises: comparability and consistency of data.

  • 1 variable: e.g. bar plot, line plot, histogram, pie plot, radar plot.
  • 2 variables: e.g. scatter plot, heatmap, map.
  • multiple variables: e.g. scatter plot matrix, mosaic plot, parallel plot.

Good overview on many techniques (starter): https://datavizproject.com/

Good overview on many tools (progressiv): https://visualisingdata.com/resources/

Visualization of Variables

Reference systems and their visual structures:

  • Temporal reference is displayed with time series: bar plot, line plot, histogram, heatmap (= techniques)

Tool: Raw Graph App and Raw Graph Turtorials

  • Spatial reference is displayed via data maps: heatmap, map, radar plot (= techniques)

Tools: https://www.mapbox.com/mapbox-studio/ https://www.datawrapper.de/maps https://hdlab.stanford.edu/palladio/

  • Abstract reference is represented by different forms of diagrams (= technique)

Tools: https://hdlab.stanford.edu/palladio/ https://app.rawgraphs.io/

  • Relational reference is represented by graphs and trees (=techniques)

Tools: https://hdlab.stanford.edu/palladio/

Examples: https://morphocode.com/location-time-urban-data-visualization/ https://humansofdata.atlan.com/2016/11/visualizing-time-series-data/

Visual structures: examples

Time Series

Number of job advertisements placed in the daily newspapers Passauer Zeitung und Kurier für Niederbayern in the period 1914–1918

Data Map

John Snow: Spatial Visualization of Cholera Cases in London 1854

Time series / Data Map

Charles Minard: Carte figurative des pertes succecives en hommes de l’Armée Française dans la campagne des Russie 1812-1813

Abstract reference: Diagram

Gantt chart: Results of the Bundestag elections 2017

Graph

Briefnetzwerk der „Fruchtbringenden Gesellschaft"

03 Tools Hands On

Workshop Website

04 Data Hands On

Data Sets

05 Workshop

Workshop Website

Literatur

  • Roxana Kath, Gary S. Schaal und Sebastian Dumm, „New Visual Hermeneutics“, in: Zeitschrift für germanistische Linguistik 43/1 (01.01.2015). Online: Crossref, DOI: 10.1515/zgl-2015-0002.
  • Malte Rehbein, Informationsvisualisierung. in: Digital Humanities: eine Einführung, herausgegeben von Fotis Jannidis, Hubertus Kohle und Malte Rehbein, S. 328-342, Stuttgart 2017.