IVDA galery

Lecture: Introduction to Interactive-Visual Data Analysis

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                                                                                                         

philipp

Further Reading

  • [VIS Design Principles]: Semiology of Graphics, Jacques Bertin, Gauthier-Villars 1967, EHESS 1998
  • [VIS Design Principles]: The Visual Display of Quantitative Information. Edward R. Tufte. Graphics Press, 1983.

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

philipp

Further Reading

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

Clara

 

Further Reading

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

Clara

 

Further Reading

  • [General Design Principles] Norman, D., The design of everyday things: Revised and expanded edition. Basic books. 2013
  • [VDA book alternative] Interactive Visual Data Analysis. Christian Tominski and Heidrun Schumann. AK Peters Visualization Series. CRC Press. 2020

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.

ieeeVIS

Further Reading

  • [IEEE VIS Best Paper] IRVINE: Using Interactive Clustering and Labeling to Analyze Correlation Patterns: A Design Study from the Manufacturing of Electrical Engines. Joscha Eirich, Jakob Bonart, Dominik Jäckle, Michael Sedlmair, Ute Schmid, Kai Fischbach, Tobias Schreck, Jürgen Bernard. IEEE VIS, TVCG, 2021.

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

Jenny

 

Further Reading

  • [Multivariate Data]: Samuel Gratzl, Alexander Lex, Nils Gehlenborg, Hanspeter Pfister, Marc Streit - LineUp: Visual Analysis of Multi-Attribute Rankings - IEEE TVCG, 2013

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

Joscha

 

Further Reading

  • Paper: Sedlmair M, Meyer M, Munzner T. Design study methodology: Reflections from the trenches and the stacks. IEEE Transactions on Visualization and Computer Graphics. 2012.

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

Joscha

 

Further Reading

  • Book: Illuminating the Path: The Research and Development Agenda for Visual Analytics. Thomas, J. and Cook, K. National Visualization and Analytics Center, 2005
  • Paper:Visual Analytics: Definition, Process, and Challenges. D Keim, G Andrienko, JD Fekete, C Gorg, J Kohlhammer, G Melancon, 2008

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)

human male 2

 

Further Reading

  • Paper: Visplause: Visual Data Quality Assessment - Arbesser, Spechtenhauser, Mühlbacher, Piringer. TVCG. 2017
  • Paper: Visual-Interactive Preprocessing of Multivariate Time Series Data - Bernard, J., Hutter, M., Reinemuth, H., Pfeifer, H., Bors, C., Kohlhammer, J. Computer Graphics Forum (CGF). 2019

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.

human male 2

 

Further Reading

  • Paper: Visual Cluster Analysis of Trajectory Data With Interactive Kohonen Maps. Schreck, T., Bernard, J., Von Landesberger, T., Kohlhammer, J. Information Visualization, Palgrave Macmillan, 2009
  • Paper: MotionExplorer: Exploratory Search in Human Motion Capture Data Based on Hierarchical Aggregation. Bernard, J., Wilhelm, N., Krüger, B., May, T., Schreck, T., Kohlhammer, J. TVCG, 2013
  • Paper: Sacha D, Kraus M, Bernard J, Behrisch M, Schreck T, Asano Y, Keim DA. Somflow: Guided exploratory cluster analysis with self-organizing maps and analytic provenance. TVCG, 2018

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.

human male 2

 

Further Reading

  • Paper: T. Mühlbacher, H. Piringer. A partition-based framework for building and validating regression models. IEEE Transactions on Visualization and Computer Graphics (TVCG). 2013
  • Paper: J. Bernard, M. Zeppelzauer, M. Sedlmair, and W. Aigner: VIAL – A Unified Process for Visual-Interactive Labeling. The Visual Computer Journal (TVCJ). 2018.

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.

human male 2

 

Further Reading

  • Book: Ben Shneiderman. Human-Centered AI. Oxford University Press. 2022
  • Paper: Bernard, J., Ritter, C., Sessler, D., Zeppelzauer, M., Kohlhammer, J., Fellner, D. Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis. IVAPP / VISIGRAPP. 2017
  • Paper: Personalized Visual-Interactive Music Classification. Christian Ritter, Christian Altenhofen, Matthias Zeppelzauer, Arjan Kuijper, Tobias Schreck, and Jürgen Bernard, EuroVA @ EuroVis (EuroGraphics). 2018