Lecture: Introduction IVDA

Lecturer	Prof. Dr. Jürgen Bernard
Teaching Language	English
Level, ECTS	MSc (3ECTS) PhD (DSI) (1ECTS)
Academic Semester	Fall 2021
Time and Location	Tuesday: 12:15 - 13:45 (starting September 28). AFL-F-121 (UZH policy: no entry without Covid-19 certificate). Zoom, for students who cannot participate in person.
Digital Backups	Slides will be on OLAT, additional course material will be as described below. There is NO RECORDING of the lecture, as this is an interactive course requiring live participation.
Start Date	28.09.2021
Exam Date	21.12.2021
Exam Location	KOL-F-101
Course Material	Coursebook (Visualization Analysis and Design, Tamara Munzner) Research papers (as announced)
Grading	Two parts, both need to be passed separately P1: 1/6 active participation across lectures, 1/6 homework1, 1/6 homework2 P2: 1/2 written exam
Office Hours	Prof. Dr. Jürgen Bernard: email for appointments, BIN 5.E.15

Course Pitch

General Description This course introduces fundamental concepts and techniques of interactive-visual data analysis (IVDA). The main focus is on the combination of automatic data analysis methods with interactive visual interfaces as well as on their interplay to facilitate data analysis goals. As such, IVDA is particularly suited to leverage the strengths of both humans and machines in a human-in-the-loop data analysis process. Associated research fields are Information Visualization, Visual Analytics, and Interactive Machine Learning.

Learning Outcome

In the first part, students will learn basic characteristics of data types and data attributes (WHAT), as well as data analysis tasks (WHY). Further, students will learn basic design skills about HOW data can be transformed into visual structures and which types of visualization techniques are meaningful design choices for given data types and analysis at hand. Students will also learn fundamental interaction techniques, as well as concepts for the composition of views in data analysis systems.

In the second part, students will gain an understanding about how data analysis can benefit from both having a human and a (machine learning) model in the loop, following the goal to gain knowledge from data. Along these lines, students will learn about the strengths (and weaknesses) of human and machines, as well as about combining these complementary strengths effectively, as described in Visual Analytics methodology. In detail, students will learn examples for interactive data preprocessing, for human-centered unsupervised machine learning, as well as for human-centered semi-supervised and supervised machine learning. Finally, the course introduces approaches that allow training personalized machine learning models and conduct personal data analytics.

Target Groups This module is designed for MA students (POC, DS). There are no enforced prerequisites. It would be possible for students in other disciplines to take this course with only minimal programming background. It is useful if students have already passed the Data Visualization Concepts lecture but once again there are no enforced prerequisites.

This course does not teach visualization libraries: most students will pick up Tableau, D3 (Javascript), ggplot (R), or python-based visualization tooling on their own.

Required Reading Visualization Analysis and Design, Tamara Munzner (A K Peters Visualization Series, CRC Press, 2014) is the course textbook. Required reading also includes selected papers as outlined below.

Topic Overview

21.09.2021		no class - start with required reading
28.09.2021	W01	Introduction to Introduction to IVDA
05.10.2021	W02	Data Types and Analysis Tasks
12.10.2021	W03	Marks, Channels, and Visualization Guidelines
19.10.2021	W04	Interaction Techniques and View Composition
26.10.2021	W05	-- no lecture -- (IEEE VIS conference)
02.11.2021	W06	Advanced Visualization Techniques
09.11.2021	W07	Users, Data Scientists, and Problem-Driven Design
16.11.2021	W08	Introduction to Visual Analytics
23.11.2021	W09	Data Transformations and Visual Preprocessing
30.11.2021	W10	ML4VIS and Data Explorers
07.12.2021	W11	VIS4ML and Model Explainers
14.12.2021	W12	Human-Centered Data Analysis
21.12.2021	W13	Exam

2021-09-21 - No class

start with the required reading (VAD book) instead

2012-09-28 - Week 1: Introduction to Introduction to Interactive Visual Data Analysis

Required Reading (pre-class)

VAD book Chapter 1. What's Vis, and Why Do It?

In-class Agenda

Welcome to the class!
Why take this course? - some considerations
Course logistics - organizational stuff
Introduction - Introduction to Interactive Visual Data Analysis
Vis-Fails

In-class Application Demo

LayoutExOmizer: Interactive Exploration and Optimization of 2D Data Layouts

Philipp Schader, Raphael Beckmann, Lukas Graner, Jürgen Bernard. VMV, Eurographics, 2021.

Presenter: Philipp Schader

2021-10-05 - Week 2: Data Types and Analysis Tasks

Required Reading (pre-class)

VAD book Chapter 2: What: Data Abstraction
VAD book Chapter 3: Why: Task Abstraction

In-class Agenda

What to visualize?: data types
Why visualize?: analysis tasks
Exercises/Questions

In-class Application Demo

Visual-interactive Exploration of Interesting Multivariate Relations in Mixed Research Data Sets

Jürgen Bernard, Martin Steiger, Sven Widmer, Hendrik Lücke-Tieke, Thorsten May, Jörn Kohlhammer. Computer Graphics Forum (CGF), 2014.

Presenter: Jürgen Bernard

2021-10-12 - Week 3: Marks, Channels, and Visualization Guidelines

Required Reading (pre-class)

VAD book Chapter 5: Marks and Channels
VAD book Chapter 6: Rules of Thumb

In-class Agenda

How to visualize? P1: marks and channels
Visualization guidelines
Exercise/Questions

In-class Application Demo

Introduction to Tableau

Tableau is an interactive data visualization tool for spreadsheet data and more. Tableau builds upon principles of the Polaris paper, with a table-based algebra for graphical presentations of tabular data.

Presenter: Clara-Maria Barth

2021-10-19 - Week 4: Interaction Techniques and View Composition

Required Reading (pre-class)

VAD book Chapter 11: Manipulate View
VAD book Chapter 12: Facet into Multiple Views
VAD book Chapter 14: Embed: Focus+Context

In-class Agenda

Hans Rosling ted talk (skipped due to technical issues)
Principles of Interaction Design
How to visualize? P2: low-level interaction techniques
How to visualize? P2: view composition and higher-level interactions
Homework introduction
Pointer: Interactive Tableau Session P2 (try it out)

In-class Video Demo (skipped)

In his video ted talk, Hans Rosling tells a interactive and visual data story about the evolution of the health situation of all states on Earth in the last 200 years. In only few minutes, he refers to 120.000 numbers. You've never seen data presented like this. With the drama and urgency of a sportscaster, statistics guru Hans Rosling debunks myths about the so-called "developing world."

Presenter: Hans Rosling (1948-2017). Global health expert and data visionary

2021-10-26 - Week 5: NO LECTURE (Reading, Homework Week)

Required Reading (pre-class)

VAD book Chapter 7: Arrange Tables
VAD book Chapter 8: Arrange Spatial Data
VAD book Chapter 9: Arrange Networks and Trees
Polaris: A System for Query, Analysis and Visualization of Multidimensional Relational Databases. Chris Stolte and Pat Hanrahan. Proceedings of IEEE InfoVis 2000. [research paper, intellectual foundation of the Tableau software]

In-class Agenda

No lecture this week (do the required reading instead)
Homework
Participate at IEEE VIS 2021 (recommended, not expected)

IEEE VIS 2021 Conference

IEEE VIS will be the year’s premier forum for advances in theory, methods, and applications of visualization and visual analytics. The conference will convene an international community of researchers and practitioners from universities, government, and industry.

We will receive the Best Paper Award for our paper called IRVINE: Using Interactive Clustering and Labeling to Analyze Correlation Patterns: A Design Study from the Manufacturing of Electrical Engines.

2021-11-02 - Week 6: Advanced Visualization Techniques

Required Reading (pre-class)

VAD book Chapter 7: Arrange Tables
VAD book Chapter 8: Arrange Spatial Data
VAD book Chapter 9: Arrange Networks and Trees
Polaris: A System for Query, Analysis and Visualization of Multidimensional Relational Databases. Chris Stolte and Pat Hanrahan. Proceedings of IEEE InfoVis 2000. [research paper, intellectual foundation of the Tableau software]

In-class Agenda

Live presentation of homework one-pagers
Overview of advanced visualization techniques
In-class Application Demo

In-class Application Demo

LineUp

LineUp is a Visual Analysis tool for the analysis of Multi-Attribute Rankings and has been published at the IEEE VIS conference in 2013. LineUp has received the IEEE VIS Best Paper Award. Students are also recommended to go to the LineUp demo page.

Presenter: Jenny Schmid

W07 - Users, Data Scientists, and Problem-Driven Design

Required Reading (pre-class)

VAD book Chapter 4: Analysis: Four Levels for Validation
Paper: Anamaria Crisan; Brittany Fiore-Gartland; Melanie Tory. Passing the Data Baton: A Retrospective Analysis on Data Science Work and Workers. IEEE Transactions on Visualization and Computer Graphics. 2021.

In-class Agenda

Data Scientists & Users
Validation of VIS Designs using the Nested Model
Design Studies
In-class Application Demo

In-class Application Demo

IRVINE

IRVINE a Visual Analytics system which facilitates the analysis of previously unknown errors in the manufacturing of electrical engines by leveraging interactive visual clustering and interactive data labeling. The design study conducted together with experts from BMW has received the IEEE VIS Best Paper Award 2021 and is published in IEEE Transactions on Visualization and Computer Graphics.

Presenter: Joscha Eirich

W8 - Introduction to Visual Analytics

Required Reading (pre-class)

Paper: Knowledge Generation Model for Visual Analytics. Dominik Sacha, Andreas Stoffel, Florian Stoffel, Bum Chul Kwon, Member, Geoffrey Ellis and Daniel A. Keim. IEEE Transactions on Visualization and Computer Graphics. 2014.

In-class Agenda

In-class Application Demo
Knowledge Generation Process
Data Transformation Processes
Humans and Machines
Synthesis: Visual Analytics

In-class Application Demo

Topic Modelling: SpecEx

Mennatallah El-Assady presents Visual Analytics solutions for complex textual document collections. Menna uses topic modeling algorithms in a text mining process, coupled with interactive visual interfaces so that humans are kept in the loop. In the demo, she shows that users' input can be useful before, while, and after an iterative topic modeling method is executed.

Presenter: Mennatallah El-Assady

W9 - Data Transformations and Visual Preprocessing

Required Reading (pre-class)

Paper: Research directions in data wrangling: Visualizations and transformations for usable and credible data. Kandel S, Heer J, Plaisant C, Kennedy J, Van Ham F, Riche NH, Weaver C, Lee B, Brodbeck D, Buono P. Information Visualization. 2011

In-class Agenda

Aspects of Dirty Data
Data Transformations
Visual Preprocessing Applied
In-class Application Demo

In-class Application Demo

Visual Analytics for preprocessing of multivariate time series

Heiko Reinemuth presents a VA system that includes a) an interactive workflow creation tool to build cascades of preprocessing operations and b) an interactive data analysis and visual comparison component to assess uncertainty and preprocessing quality. Co-authors of the EuroVis conference and CGF journal publication are J. Bernard, M. Hutter, H. Reinemuth, H. Pfeifer, C. Bors, and J. Kohlhammer.

Presenter: Heiko Reinemuth (former student and co-author of Prof. Bernard)

W10 - ML4VIS and Data Explorers

Required Reading (pre-class)

VAD book Chapter 13: Reduce Items and Attributes

In-class Agenda

"ML4VIS" with an emphasis on data-centered VIS challenges
Visual Analytics based on Clustering
Visual Analytics based on Dimensionality Reduction
Self-organizing Maps (SOM) special
In-class Application Demo

In-class Application Demo

SOMFlow: exploratory cluster analysis with Self-Organizing Maps (SOM)

Dominik Sacha presents a multi-stage VA approach for iterative cluster refinement using SOMs to analyze time series data. SOMFlow is a visual platform to analyze intermediate results, adapt the underlying computations, iteratively partition the data, and to reflect previous analytical activities. Co-authors of the TVCG journal publication (2018) are Sacha, D., Kraus, M., Bernard, J., Behrisch, M., Schreck, T., Asano, Y., Keim, D.

Presenter: Dr. Dominik Sacha(Siemens), former PhD student at the University of Konstanz.

W11 - VIS4ML and Model Explainers

Required Reading (pre-class)

Paper: Hohman F, Kahng M, Pienta R, Chau DH. Visual analytics in deep learning: An interrogative survey for the next frontiers. IEEE Transactions on Visualization and Computer Graphics (TVCG). 2018.

In-class Agenda

Supervised ML Special
VIS4ML - Using VIS/VA to support ML
Live Application - VIAL – Visual Interactive Data Labeling
In-class Application Demo

In-class Application Demo

Visual Observation of Data Labeling Runs

Raphael Beckmann presents an interactive visual analysis tool for the (retrospective) observation of data labeling runs. The approach provides multiple linked views for the visual assessment of data labeling quality and the choices the applied instance selection strategy.

Presenter: Raphael Beckmann is a student member of the IVDA group.

W12 - Human-Centered Data Analysis

Required Reading (pre-class)

none

In-class Agenda

Human-Centered Artificial Intelligence (HCAI)
Introduction to Human-Centered Data Analysis
Types of Explicit User Feedback
In-class Application Demo

In-class Application Demo

Personalized Similarity and Personalized Classification

Christian Ritter presents two human-centered data analysis tools. The first allows users to express their personal notion on the similarity of soccer players, the second supports users in labeling personal music collections. In both cases a machine learning model is trained iteratively that can also be analyzed interactively. The Soccer and the Music tool have both been published in the visual analytics community (see Further Reading below).

Presenter: Christian Ritter is a former student and co-author of Prof. Bernard.

Course Pitch

Topic Overview

2021-09-21 - No class

2012-09-28 - Week 1: Introduction to Introduction to Interactive Visual Data Analysis

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

2021-10-05 - Week 2: Data Types and Analysis Tasks

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

2021-10-12 - Week 3: Marks, Channels, and Visualization Guidelines

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

2021-10-19 - Week 4: Interaction Techniques and View Composition

Required Reading (pre-class)

In-class Agenda

In-class Video Demo (skipped)

Further Reading

2021-10-26 - Week 5: NO LECTURE (Reading, Homework Week)

Required Reading (pre-class)

In-class Agenda

IEEE VIS 2021 Conference

Further Reading

2021-11-02 - Week 6: Advanced Visualization Techniques

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

W07 - Users, Data Scientists, and Problem-Driven Design

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

W8 - Introduction to Visual Analytics

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

W9 - Data Transformations and Visual Preprocessing

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

W10 - ML4VIS and Data Explorers

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

W11 - VIS4ML and Model Explainers

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

W12 - Human-Centered Data Analysis

Required Reading (pre-class)

In-class Agenda

In-class Application Demo

Further Reading

Fusszeile