Visualisation

From WebScience

Jump to: navigation, search


Fact Box
Module Foundations and Principles I
Course
representative
Martin Eisemann
Credits 3
Term Term 1, Term 2
Course is not required
Current course page Winter 2018
Active Yes


The Big Picture

The history of visualisation dates back to the stone age when hunters carved their story into the cave walls to provide information on how to capture prey in order to survive. Nowadays, information is still as important, think of Facebook, Google, and co. The amount of information produced in science, engineering, business or any other area is tremendously increasing and it becomes ever more important to make this information available, readable, and understandable not only for a computer but especially for us as humans.

Visualisations do not only present a visual representation of the data but they help to improve comprehension and decision making, communicate a story, entertain, and they can lie! In this course you will become aware of the intricacies of visualisations. We will investigate commonly acknowledged design principles, subjective dimensions that make a visualisation appealing to the viewer, and take a look at good and bad design practices. You will learn about different methods for visualisation of data from various fields, about what kind of data exists and how data can be represented as a visual variable. During the course you will learn to evaluate and have a critical view on visualisations and create and review your own visualisation.

Intended Learning Outcomes

After participation in the course, the students are able to analyze and discuss visualisations from a critical perspective. They are able to

  • evaluate, review and critically assess visualization designs
  • apply fundamental principles & techniques
  • design interactive visualizations by developing a substantial visualisation project
  • communicate their findings to others

Structure of the Course

Introduction

Starting with a brief history on visualisation and an introduction to this issue, some basic terms and concepts are defined[1][2][3][4][5]

Design principles

Based on the work of Tufte[2][4] we will learn about design principles, how to evaluate and critique visualisations and how to decipher the intended audience and questions answered by visualisations. We will look into graphical integrity, in contrast to the rise of fake news. We will also investigate the use of color and subjective dimensions and its purpose in visualisations.

Data model

The data model describes the type of data (nominal, ordinal, and quantitative) and which operations can be applied to it. [6]

We'll discuss representations of data and their relation (Entities, Relationships, Attributes) and how to map the data model to the conceptual model and vice versa.[1]

Visual variables

In visualisation data needs to be mapped to visual variables representing the different dimensions. We'll learn about how these mappings can be conducted and which variables are most distinct to us as humans[7] These include e.g. position, size, grey value, texture, color, orientation, and shape.

Didactic Concept, Schedule and Assignments

The course concept comprises basic readings, online research, workshops, group work, and lectures. After a first introductory lecture on site, the organization of the course is introduced and the subject is treated in three online workshops. Each of the three online workshops has a duration of three hours each. In a final session on site the results of the conducted project will be presented in a poster session.

Taking an active part in this course is crucial as the didactic concept is built on interaction and discussion between the students and between students and the lecturer.

The online workshops will be divided into lectures, group works, and discussions in the plenum. During the group works the students will learn to evaluate, critique, and create various visualisations.

For each workshop an assignment has to be completed by each of the participants beforehand which will provide essential content needed during the workshops. Therefore, the participants are expected to have their solutions to the assignments ready for the respective workshops.

Introductory lecture on site

In the introductory meeting we will motivate the topic of visualisation and its importance with a short introductory lecture[2] [1]. We then deal with the organizational components of the course.

1st assignment

  • Search the internet for 3 interactive visualisations you like.
  • For each of them write down
    • what they are used for,
    • what you like about them,
    • who the intended audience is (children, data experts, farmers, etc.), and
    • which questions the visualisation answers.
    • Write down any suggestions to improve the visualisation further.


1st online workshop

In this workshop we will investigate several of the collected interactive visualisations and discuss their goals, qualities, and intended audience. For this we will introduce Tufte’s Design Principles and Design Critique in a short lecture of about 1h. The participants will then be partitioned into small teams. In each team

  • every member will shortly introduce his/her found visualisations and introduce it to the other members in 2-3 minutes per visualisation.
  • the members will quickly reason about the value of the visualisation based on Tufte’s Design Priniciples.
  • the team members choose the best and worst visualisations, based on all the presented visualisations.

This part of the workshop should take about 1h.

In the last part each team will present its best and worst visualisation to the other participants of this course and we will discuss the benefits and drawbacks in the plenum. (1h)

2nd assignment

  • Read the paper „How to Read a Visualization Research Paper“[8] available here[9].
  • Go to the IEEE Vis 2017 homepage[10] or any other year you like and go to the fast forward video collection[11] Take a look at several of the videos which represent the current state-of-the-art in visualisation research. Try to gather and structure the videos and their respecting publications according to the following:
    • What common characteristics does the data used in these visualisations have?
    • Do you encounter reoccurring types of visualisations? What are these?

Note: You can access all articles via IEEE Xplore[12] but you need to create a VPN connection to the TH Köln to access the papers.

  • Count and write down information about all the dogs you meet during the week. This is no joke, we will need this information. Think about as many aspects as possible which you could gather, e.g. time, size, color, race, if it barked, etc.

2nd online workshop

In this workshop you will learn a bit more about common visualisation techniques used for exploratory analysis. In addition you will create your own visualisation technique, so bring along pen&paper (and something to digitalize the result) or digital drawing tools.

The participants will be partitioned into small teams. In each team

  • start clarifying any questions about Laramee’s paper.
  • each team member should then shortly introduce the visualisation paper/video he/she liked best from the fast forward videos. Discuss the drawbacks and benefits.
  • gather the information you collected about the reoccurring visualisation types.

This part of the workshop should take about 45 minutes.

The workshop then continues with a lecture amending any missing common data types, visual variables, or typical visualisations used in information visualisation and visual analytics, as well as how to map data to visual variables. This will take about 45 minutes.

In the last part of the workshop you will gather in small groups again,

  • combine the dog data each group member collected and
  • create designs for visualising them, e.g., using pen&paper. You should create at least 2-3 different designs,
  • discuss its drawbacks and benefits and vote on them to select the best.
  • Create examples based on your data set and also show how extreme cases would look like.

Make sure that your design thoroughly describes what data dimensions you have and how these are mapped to the visualisation. This part should take about 1h.

Finally, each group will present their design to the rest of the course. The presented designs must not contain any text except to describe the dimensions represented in your visualisation! The data must be visually represented!

3rd assignment

  • Look for tools on the internet that let you create visualisations (to get you started look e.g. for D3, Python, R, Tableau)
  • Collect at least three different tools from which at least two mustn’t be in the aforementioned list.
  • Investigate the possibilities, intended user, required knowledge, and limitations of these and write them down.
  • Collect at least one minimal visualisation example from the internet to get a feeling for how visualisations are created with these tools.
  • Collect ideas and data sources for your final project from the internet. Where could you get data from? How much effort would it be to get the data and transform it into a usable format, e.g., CSV-files.

3rd online workshop

In this workshop we will discuss the collected tools and begin to define the final projects which have to be presented at the wrap-up session on site. In the plenum we will first gather the collected tools. (10min)

Based on the selected list we will create small groups of 2-3 people where each group discusses 1-2 tools. (20min)

The outcome of this discussion will be presented in the plenum. (15min)

In the second phase, the participants will be partitioned into small teams of 2-3 people for the final project. In your team

  • discuss the collected project ideas. Every team member should quickly present his/her ideas to the others in about 1-2 minutes per idea. Discuss the ideas. Pay special attention to potential, creativity, availability of data sources, and implementability. (20min)
  • vote on the projects. Select the one that your team finds most interesting to work on. (5min)
  • develop ideas which questions should be answered by the visualisation and who should be the intended audience. (10min)
  • For each of these questions start designing visualisation mock-ups, e.g. pen&paper prototypes that convey the idea of the visualisation. (1h)
  • Decide on the tools you plan to use to implement your project. Can these tools provide everything you need? (20min)

Finally, each group will present their idea and mock-ups to the rest of the course. The presented designs must not contain any text except to describe the dimensions represented in your visualisation! The idea must be visually represented. (20min)

Project

You will need to prepare three things in your final project: A project plan, a project, and a poster. All three components (project plan, project files, poster) must be uploaded to the web science wikipage of this course no later than three days in advance to the wrap-up session on site.

Project plan

Based on your project mock-ups, you will create a detailed project plan, which should address the following points.

  • Basic Info. The project title, your names, e-mail addresses, student number, a link to the project URL (this link should provide all files of your project).
  • Background and Motivation. Discuss your motivations and reasons for choosing this project, especially any background or research interests that may have influenced your decision.
  • Related Work. Anything that inspired you, such as a paper, a web site, visualisations we discussed in class, etc.
  • Project Objectives and Goals. Provide the primary questions you are trying to answer with your visualisation. What would you like to learn and accomplish? List the benefits.
  • Tasks. Describe in detail which data manipulations (sort, filter,..) and visual manipulations (zoom, selection,…) you would want to implement and how these support the goals.
  • Data. From where and how are you collecting your data? If appropriate, provide a link to your data sources.
  • Data Processing. Do you expect to do substantial data cleanup? What quantities do you plan to derive from your data? How would data processing be implemented?
  • Visualisation Design. How will you display your data? Provide some general ideas that you have for the visualisation design. Develop three alternative prototype designs for your visualisation. Create one final design that incorporates the best of your three designs. Describe your designs and justify your choices of visual encodings. We recommend you use the Five Design Sheet Methodology (fds.design).
  • Must-Have Features. List the features without which you would consider your project to be a failure.
  • Optional Features. List the features which you consider to be nice to have, but not critical.
  • Project Schedule. Make sure that you plan your work so that you can avoid a big rush right before the final project deadline, and delegate different modules and responsibilities among your team members. Write your schedule in terms of weekly deadlines.
  • Implementation details. List what tools/frameworks you used to implement your interactive visualisation. List the difficulties you encountered. List which of the desired features are implemented and which are missing.
  • Work Breakdown Structure (e.g. matrix) with a statement which project member did which part of the structure to what extend.

As a ballpark number: your project plan should contain about 3-4 pages of text, plus 5-6 pages of sketches and images of the final visualisation with explanation. Make sure to list a link to the final visualisation.

Final visualisation

You are free to choose any format of representation for your final visualization. This can range from pen&paper-prototypes, over Power Point presentations, to usage of existing tools, such as Tableau, up to to real implementations as a web page or program. Choose the method you feel most comfortable with and with which you expect the best results.
 Part of the grading will incorporate the amount of work required to create the prototype.

If you choose to create a real implementation, make sure you cite all sources that you use. This is mandatory.

If you choose not to create a real implementation (maybe because you have no experience in programming), you can use made up data to convey your idea. Your visualisation should nevertheless be interactive, e.g. a clickable PowerPoint presentation or goto-assignments for pen&paper prototypes.

For grading the final project, we will evaluate projects by the following criteria:

  • Effective visualizations
  • Innovative visualizations
  • Level of technical difficulty
  • Clear storytelling
  • Visual design
  • Addresses the goals
  • Sensible and effective interaction

Make sure all intended interaction is represented in your prototype or at least in the design mock-ups.

Poster

Also create a DIN A0 poster describing the gist of your visualisation as a PDF which we will use for the wrap-up session on site. Upload the poster as a PDF to the web science wikipage at least three days before the on-site meeting.

Wrap-up session on site

In the wrap-up session on site we will create a poster session. We will print the posters for you. From each team two people have to present its project in front of their posters while the others investigate the presentations of the other teams. Please cycle through your team to change the presenters from time to time. Make sure your presentation is short enough so that each course member can take a look at all posters.

Each team has to present its interactive prototype. For this you should bring your own laptop or PC if required. In case your team does not have a laptop or PC, send your visualisation to your lecturer at least one week in advance with a detailed installation description. In case you created a web visualisation, create an appropriate web server and send the lecturer your link if you don’t bring along your own laptop.

Examination

  • 50% of your grades will be determined by your project plan.
  • 30% of your grades will be determined by your final project itself.
  • 20% of your grades will be determined by your poster.

The grades will be reported to the examination office shortly after the on-site weekend (~1-2 weeks later). You can get feedback to your grading if required by contacting your lecturer.


References

  1. 1.0 1.1 1.2 Matthew O. Ward and Georges Grinstein and Daniel Keim (2015). Interactive Data Visualization (2nd Edition). CRC Press. 
  2. 2.0 2.1 2.2 Edward Rolf Tufte (2001). Visual Display of Quantitative Information. Bertrams. 
  3. Ben Fry. Visualizing Data. O'Reilly. 
  4. 4.0 4.1 Edward Rolf Tufte (1990). Envisioning Information. Graphics Press. 
  5. Alexandra C. Telea (2014). Data Visualization Principles and Practice. Taylor & Francis Inc. 
  6. Stevens, S. S. (1946). "On the Theory of Scales of Measurement". Science. New Series 103 (2684): 677--680. 
  7. Jacques Bertin (1983). Semiology of Graphics. University of Wisconsin Press. 
  8. Robert S. Laramee (2011). "How to Read a Visualization Research Paper: Extracting the Essentials". IEEE computer graphics and applications (IEEE) 31 (3).  by Robert S. Laramee
  9. "How to Read a Visualization Research Paper: Extracting the Essentials". https://cs.swan.ac.uk/~csbob/research/how2read/laramee09how2read.pdf. Retrieved 17 August 2018. 
  10. "IEEE Vis 2018". http://ieeevis.org/year/2017/welcome. Retrieved 17 August 2018. 
  11. "IEEE Vis Fast Forward Videos". https://vimeo.com/groups/480818. Retrieved 17 August 2018. 
  12. "IEEE Xplore". https://ieeexplore.ieee.org/. Retrieved 08 October 2018. 


Past Course Pages