Karteikarten
Data Visualization
Karten in diesem Stapel (35)
- 1THE VISUALIZATION PIPELINE
- 2WHAT IS DATA VISUALIZATION?Both a product and a process
- 3What is the composition of color?Hue = Wavelength
Saturation = Intensity from gray to pure, vivid color
Brightness = Lightness from black to colorful to white
- 4What is LuminanceVisually perceived brightness, not uniformed defined
- 5Name three GESTALT PRINCIPLES OF GROUPING
- 6What are the four data types
- 7What is EXPRESSIVENESS?Showing the data and only the data
- 8How to consider EXPRESSIVENESS?A visualization is effective if it exploits the ccapabilities of the output
medium and the human visual system
- 9What does Appropriateness mean for a data viz.A visualization is appropriate if it balances the tradeoff between efforts
required for creating it and the benefits yielded by it. - 10What are the two main ways we can design the interface for the viz?OVERVIEW + DETAIL and FOCUS + CONTEXT
- 11When and how should we use multiple coordinated views ?Baldonado rules:
Rule of Diversity: Use MCV in case of many attributes, purposes, users,…
• Rule of Complementarity: Use MCV to show different aspects of the data
• Rule of Decomposition: Use MCV to deconstruct complex data into manageable / understandable parts
• Rule of Parsimony: Don’t use MCV when a single view suffices
• Rule of Space/Time Resource Optimization: The resources saved by MCV should outweigh their costs.
• Rule of Self-Evidence: Relations between views should be clear.
• Rule of Consistency: Use consistent interfaces and encodings across MCV
• Rule of Attention Management: Guide the user focus to the right view at
the right time. - 12What is DATA CONTEXT and DATA CONTENTThe data context give us the identity of the data we are looking at, while the data content is the part of the data there is measured.
- 13What is a task?A task can be seen as a mid-level abstraction between the goal of the user and the action they want to take. IT can be sometinh like a overview or Zoom+Filter. There are three catagories, exploration, insight and meta.
- 14What is the difference between s Elementary task and a synoptic tasksElementary tasks are tasks that ask/answer questions
about individual data items and their relations, while synoptic tasks are tasks that ask/answer questions
about sets of data items and their relations - 15What is the DESIGN ACTIVITY FRAMEWORK made up of ?Understand: Motivation: to gather and research the needs of the use.
Outcome: visualization requirements
Ideate: Motivation: to generate good ideas for supporting the requirements
Motivation: to generate good ideas for supporting the requirements
Make:
Outcome: a testable hi-fi prototype
Deploy:
Motivation: to bring a prototype into effective action in a real-world setting
Outcome: a testable hi-fi prototype - 16What is The 5 Design Sheet Methodology?IT is a method to come up with vis ideas.
It contains five sheet. THe first sheet is used to come up wiht many samll ideas and the combine then into a few big ones.
Sheet 2 to 4 is then used to fill out three different ideas and sheet 5 is used to clarify which one we pick with more details. . - 17What are the GUIDING DESIGN PRINCIPLES ?• Functional Design:
-> First create something useful, then make it more beautiful.
• Deliberate Design:
-> Don’t include any visual feature without justification (purpose or reason).
• Intuitive Design:
-> The visualization should not be more complex than the underlying data.
• Ethical Design:
-> Do not visualize to deceive the reader/viewer - 18What is MUNZNER’S NESTED MODEl
- 19What is the three stages of Data PreprocessingData Profiling, Data Wrangling and Data transformation
- 20How can we handle missing data?Using a placeholder value (bad), data imputation , data amputation and data imputation with data imputation.
- 21What is common parts of data transformation?DATA REDUCTION by random sampling or some other sampling
AGGREGATING DATA by clustering using different linkage metrics - 22How can we deal with multivariate data with points?• Triliniar Plots = Scatterplots for 3 variables that add up to a given total
• Scatterplot Matrices (SPLOMs) = One Scatterplot per variable
combination in a matrix layou
• RadViz = each attribute is assigned
an anchor on the circumference
of a circular drawing area - 23How can we deal with multivariate data with lines?PARALLEL COORDINATE Plots(PCS)
RADAR CHART - 24How can we deal with multivariate data with Nesting?Trellis Displays = make a grid where each entry coves some combination of variables.
Mosaic Plots
- 25When mapping time to visulas which are the most effective choices?Position, lenght and angle.
Lines with. brightness and connection can also be used.
Color can be fine if other option have been taken - 26What graph is good at comparing multiple timelines?Horizon Graphs:
- 27Why can it be hard to make good GEOVISUALIZATION?We cannot use positions
and the Modifiable areal unit problem which comes from aggregating point-based data into areas - 28What do we want from a map projection?Preserving angles: mercator projection
Preserving areas:
Mollweide projection
Preserving distances:
Werner projection
We will need to chooce - 29How can we add data to a map?Points: Making either a dotmap or a heatmap
Lines: FLow map, edge-nundled map
Areas: Choroplth map, isoplet map
Multivar: Glyphs, Trellis plot
- 30How can we adjust the map to fit the data?Area/ Continuous Cartograms
Dorling Cartogram
Demers Cartogram
The RecMap Algorithm
Gridded / Mosaic Cartograms - 31WHAT MAKES A GOOD NODE-LINK-DIAG.?• Draw edges straight, polyline,
orthogonally,…
• Minimize Edge Crossings
• Minimize Edge Bends
• Minimize Aspect Ratio
• Minimize Drawing Area
• Maximize Symmetries
• Maintain Uniform Edge Length
- 32What makes up the GRAPH VISUALIZATION PIPELINE?• Node Positioning (Graph Layout)
• Graph pre-processing
• Pre-layout
• Layout
• Post-layout
• Edge Routing
• Edge Bundling
• Attributes -> encode
• Labels
• Colors - 33WHY USE MATRIX VISUALIZATIONS?1) It allow for a compact representation of dense subgraphs
2) For egde-centric displays
- 34What is VISUAL ANALYTICS?Visual Analytics denotes the seamlessly
integrated data analysis using
automated, computational methods and interactibe graphical methods
- 35What is the design space?
Informationen
- Beschreibung
- Data Visualization Karteikarten am Data Visualization, erstellt von Simon Lauridsen am 05/01/2022.
- Erstellt von
- Simon Lauridsen
- Aufrufe
- 6
- Bewertung