Stresa, Italy
September 5-7, 2001

Tutorial 1

Information Visualization
Bob Spence (Imperial College, London)

Synopsis

Many people and institutions possess considerable volumes of data which may 'hide' some fundamental relation which could be exploited to advantage: estate agents, banks, medical researchers, fraud investigators and many others would like to be able to view some graphical presentation of that data and perhaps interact with it and, at some point, be able to say "Ah Ha! Now that is interesting!" That is what Information Visualization is all about: it is the process of forming a mental model of data, thereby supporting insight into that data.

Tutorial organisation

The tutorial is organised into 7 sections based on a reasonable taxonomy of the subject. Times are approximate, and can easily be rearranged to suit planned coffee breaks.

The Issues: A brief historical review providing static examples [London Underground; Minard's map; Florence Nightingales 'roses'; Dr. John Snow and cholera, etc] that help to establish issues to be treated in subsequent sessions involving the application of computers to visualization [Selection; Representation; Presentation; Scale and Dimensionality; Rearrangement, interaction and exploration; data externalization] (25 mins)

Interaction, Rearrangement and Responsiveness: Why rearrangement of visually presented data is needed [examples: crop teatments, supply distribution networks, hifi design, landscape presentation]; the design of affordances [examples from conventional doors, Table Lens, scrolling control], and the benefits of responsiveness (40 mins)

Representation of data: The many ways in which data (whether numeric, ordinal or categorical) can be presented, appropriate to the task being performed by a user. Examples include Box Plots; histograms; scatterplots and matrices; brushing; 3D presentation; Parallel Coordinate Plots; Mosaic Plots and a variety of encoding techniques including size, magnification, colour, Chernoff faces, multidimensional icons, patterns and encoding by sound. Demonstration of Parallel Coordinate Explorer, and video clips of visualization tools. (45 mins).

Selection of objects: Many visualisation tools facilitate the task of selecting one object from among many. Dynamic Queries, Attribute Explorer, EZChooser, Neighbourhood Explorer, Model maker. Demonstration of Attribute Explorer, and video clip demonstrations. (45 mins)

Presentation: Too much data - too little display space: solutions to this problem. Bifocal Display, Perspective Wall, Suppression, RSVP.(Rapid Serial Visual Presentation) and others. Application to mobile and PDA displays. Video clip demonstrations. (40 mins)

Connectivity: Users often wish to gain insight into things that are connected. Fraud investigation using Netmap, and, for tree-structured data, Hyperbolic Browser, Cone trees and Treemaps. Video clip demonstrations. (30 mins)

Models: In some important situations such as design data is not available a priori, but is generated as required to explore otherwise hidden relations ships. A brief discussion, supported by a demonstration and video clips, based on engineering design (20 mins)

As noted, the course will be illustrated by a number of demonstrations of visualization tools (Parallel Coordinate Plots, Attribute Explorer) as well as many short (10 secs to 5 min) video clips.

Who can benefit from this tutorial

Anyone who possesses data into which they or their colleagues need to gain insight. CEOs, software engineers, web designers, marketing directors and scientists: the list is long. It is important to make it clear to potential attendees that no knowledge of maths, computing or statistics is required

About the speaker

Bob Spence is Professor of Information Engineering at Imperial College, London. He began his work in Human-Computer Interaction in 1967 with work on the interactive-graphic design of circuits, work which eventually led to a commercial CAD tool in 1985 marketed by a company co-founded and chaired by him. Other research and innovation since then included many novel techniques for information visualisation such as the Attribute Explorer, the Influence Explorer and the Prosection Matrix. With a colleague he was responsible for the invention of the Bifocal Display, now often referred to as the Fisheye Lens, and is currently involved in research into a new method of data display called Rapid Serial Visual Presentation (RSVP) having particular relevance to small displays. For his work in Human-Computer Interaction he was awarded the higher doctorate of the Royal College of Art. Concurrently with his work on Human-Computer Interaction, Professor Spence is widely known for his engineering research, and particularly for methods of enhancing the manufacturing yield of mass-produced artefacts. He is a Fellow of the Royal Academy of Engineering.

Based on his recently published book Information Visualization (Addison-Wesley) Professor Spence has previously presented his course on Information Visualization at HCI'2000, in Sydney and Melbourne in early 2001, in Detroit (30/3/01), Vancouver (6/4/01) and is due to present it again in London (IV'2001, July) and in France in September. He has also taught a 2-week module on the same subject at the Technical University of Eindhoven in 2000, to be repeated in April/May 2001.

Last update: May 8, 2001