Quick Links

GAV
GAV API
Forum
Slides & Papers (2009)
Slides & Papers (2010)
Slides & Papers (2011)
Slides & Papers (2012)

Information Visualization

Forum Information

To be able to post in the forum you have to join the group "TNM048 Students". First create an account, then go to "User Control Panel" (Below "Board Index"), click "User groups", then select and join the appropriate group.

Examinations

Course

The course integrates perspectives from multiple fields, including information visualization (InfoVis), visual data mining, exploratory data analysis (EDA) and geographical visualization (GeoVis). These domains are often referred to the name Visual Analytics http://nvac.pnl.gov/. InfoVis is a relatively new discipline but already indispensable as a provider of tools for accessing and managing large (and rapidly growing) often abstract information spaces. Visual data mining and visual knowledge discovery seem to emerge and establish themselves as sub-disciplines. InfoVis research has a strong bearing on current and future human-computer interfaces. InfoVis is about presenting abstract data in visual form to make it more understandable - to amplify cognition. The purpose can be the presentation of the data to others or the use of the visualization methods to study the data and to aid decision-making

The modern computing society also has to deal with a geographical or spatial component. Visualization of such data traditionally belongs to the research domain known as geographic visualization, geovisualization or the new domain name Geovisual Analytics http://geoanalytics.net/. Spatial data often have a complex structure involving space, time, and a number of demographics attributes. In GeoVis applications, such data are represented using both traditional cartographic techniques based on the use of colours, textures, symbols, and glyphs. Moreover, maps are used in combination with InfoVis techniques such as scatter plots or parallel coordinates. The use of multiple interactively linked views providing different perspectives into the data has become a kind of standard in GeoVis.

InfoVis is core to innovation and insight and of strategic industrial and public service importance as well as indisputable scientific value. InfoVis offers the potential to provide managers, analysts and experts in academia, industry and public services with competitive decision-making tools to:

The overall objectives of the course is to learn basic visualization and interaction techniques in the information and geovisualization fields, as well as basic approaches to visually exploring large databases and visual data mining.

Read more: http://vita.itn.liu.se/research/information-and-geo-visualization or our new web site http://ncva.itn.liu.se.

This is the home page for the course Information Visualization (TNM048) and is currently located at http://vitagate.itn.liu.se/courses/TNM048/. It contains current schedule and information about the course. All material distributed during the course will be linked to from this page (or there will be a description on where to find it). This page will be updated during the course.

Why should I take this course?

Information is everywhere. In computer science, visualization can be used in network security, algorithm visualization, or as an enhancement to data mining. In other fields, anywhere where there is a large amount of computer readable spatial data, InfoVis and GeoVis can be used to understand it.

Examination Form

For achieving grades (6p) in the InfoViz TNM048 course:

1. Accepted lab assignments: Lab1 and Lab2 (InfoVis) and Lab3 (GeoVis). The in-house (NVIS) developed toolkit GeoAnalytics Visualization (GAV) framework will be used as the programming framework. GAV is a low-level component toolkit based on InfoVis, GeoVis and even some SciVis methods. The toolkit is developed in C# and use Microsoft DirectX graphics library. The labs will give you experience with designing and assembling InfoVis and GeoVis applications in Microsoft?s .NET, C# and DirectX environment. (1,5p)

2. Accepted written test based on contents (slides etc.) from lectures. (1,5p)

3. Accepted group assignment (3p).

Read more about the GAV at: http://vita.itn.liu.se/GAV
Code related resouces are available at http://vitagate.itn.liu.se/projects/GAV/framework/

Lectures

Information Visualization (InfoVis) is an important often applied research topic and can be described as the use of computer-supported interactive visual representations of abstract data to amplify cognition. Whereas scientific visualization usually starts with a natural physical representation, InfoVis applies visual processing to abstract information. This area arises because of trends in technology and information scale. Technically, there has been great progress in high-performance, affordable computer graphics. At the same time, there has been a rapid expansion in on-line information, creating a need for computer-aid in finding and understanding them. InfoVis is a form of external cognition, using resources in the world outside the mind to amplify what the mind can do.

Progress in InfoVis has occurred along four significant dimensions. First, the original ideas have matured. InfoVis captures the themes from early work by extending SciVis which focus on 3D scientific data toward understanding non-spatial, higher-dimension data. Second, the research focus has grown beyond visualization to include data mining and analytics as integrated components of the knowledge discovery process. Third, the applications have grown in number, depth, significance, and sophistication. Fourth, InfoVis has the potential of engaging the human mind in creative analytical interpretations that can lead to the discovery of important insight. Users with diverse background and expertise could participate in such a creative discovery process. Recognized scenarios, however, that cannot be captured remains within the mind of a single user and is not easily accessible to external analyze. Recent InfoVis research now also provides methods that integrate the visual analytics process with collaborative means and can streamline this knowledge exchange process of developing a shared understanding with other users.

In the course, you will gain insight and practise into the following InfoViz and GeoViz techniques:

A special invited lecture is about large, multivariate data visualization is given. This lecture will cover data- and screen-space methods for clustering, transformation and dimensionality reduction. Examples of methods that will be discussed in detail are: K-means, self-organizing maps and density maps.

A second invited lecture covers categorical data. Different data types have different properties. One type of data that involves particular difficulties in visualization and hence need special treatment is categorical data. You will be introduced to the most common data types in information visualization, and will gain knowledge of the difficulties concerning categorical data and of the existing ways of visualizing categorical data. Furthermore you will be introduced to some recent research in the area of categorical data visualization.

The course will also address the closely linked field GeoVis. GeoVis is a process that meets scientific and societal needs by providing visual methods and tools to support a wide array of geospatial data applications including real-world knowledge construction and decision-making.

Example of application areas for InfoVis and GeoVis:

The InfoVis part of the course includes 5 lectures (4 lectures on InfoVis, 1 on GeoVis) by the course leader professor Mikael Jern (mikael.jern@liu.se).

Camilla Forsell, postdoc, (Camilla.Forsell@liu.se) will give a lecture about application evaluation.

Jimmy Johansson, PhD (jimmy.johansson@liu.se) will give an invited lecture about large multivariate data.

Slides

All slides can be found here

Labs

The programming part of the course (Lab1, Lab2 and Lab3) will cover the basics of the GAV components to provide the practical skills needed to experiment with the methods covered in the lectures. The lab assignments will be based on GAV and is divided into three parts. These courses and labs will give you a unique introduction to development of state-of-the-art InfoViz applications in .NET environment. The lab assistants have gained much experience in programming application components and will give the training courses, lead the labs and provide support.

The labs will be carried out in groups of 2 students.

All the material for each lab comes in the .zip files below. Instructions are in pdf format and any code will be in projects compatible with Visual C# express 2008. The labs are the same as 2009, as indicated by the filename.

Lab and Project Assistance

Patrik Lundblad (patrik.lundblad@liu.se), PhD student, and Quan Ho, PhD Student, (ho.van.quan@liu.se) will provide assistance in the labs and project assignments. They have their offices in Spetsen on 5th floor.

Project Assistance

Each project group will have the possibility to sign up for 2x30 min of project supervision with one of the lab assistants. A schedule of available supervision times and a list to sign up for supervision will be available as soon as we know how many groups there will be. More information about this will come later on.
Remember to use the forum aswell as email.

Visualisation Tools

The visualization components that will be used in the course are based on our in-house developed GAV Framework: http://vitagate.itn.liu.se/projects/GAV/framework/), a development platform that provides low-level InfoVis and GeoVis components for developing customized visualization applications. For specific GAV related questions or bug/feature fixes, contact Tobias Åström (tobias.astrom@liu.se).

Forum for GAV questions

Last year students suggested that we add a forum for questions and answers regarding developing using GAV. The idea is to post all questions that surface during the course to take the load of the lab assistants, and to make it easier for students to get their questions answered. The forum will also contain posts regarding webpage updates and bug-fixes to GAV.
To the forum

Group Assignments

Project Assignment 2010 - Full information (.pdf)

The group assignment will be carried out in groups of 2 people that will solve an exercise by developing an integrated InfoVis and GeoVis application using a map with attribute data for Sweden (municipalities), World or Europe (TL2 or TL3) that meets the given demands. The work will be documented in a detailed report that contains thorough explanations and motivations to used methods and visualization options.

Assistance will be given by Quan Ho, PhD Student (ho.van.quan@liu.se), Patrik Lundblad (patrik.lundblad@liu.se), PhD student.

A set of InfoVis and GeoVis problems related to specific datasets public available in, for example, Sweden Statistics Database (http://www.scb.se/) or OECD will be used. Each group can choose between one of the data problems to work on. The group decides which tools and techniques to use to make a visualization that fulfils the group assignments goals. (GeoAnalytics, .NET, DirectX, HTML, Flash, OpenGL, VR etc.)

Accepted group assignment includes the following four tasks, which must be accepted (percentages of contribution to final grad):

The sense making loop

All four tasks will be individually judged and result in a joined grade (3-5) or unacceptable. All participants in a group will get the same grade.

Example of Project Reports (.pdf)

Example 1
Example 2
Example 3
Example 4

Example of Data sets

The data should preferable contain 3 dimensions: spatial, time and multivariate attributes.
  1. Swedish census data: SCB Census
  2. Swedish Energy data: SCB Energy
  3. World Data: Earth Trends, World Bank, International Monetary Fund
  4. European Data: Eurostat
  5. SKL - Sveriges Kommuner och Landsting
Please, note that most text information about Swedish demographics data is described in Swedish. We recommend that our foreign master students collaborate with a Swedish speaking student or use data about Europe or the world.

Basic Example Component

SampleComponent.cs + Glyph texture
Building a layered component

Code resources

MSDN Library: C# Center
Good DirectX & HLSL tutorials in c#

Project Presentations

You may select to make your project presentation on a few occasions that will be presented here at a later date. A list will be available where you can sign up in Spetsen Floor 5.

Literature and other resources

Course literature will be the lecture slides and any given references to web sites. The lecture slides will be provided as a PDF document after each lecture.

Recommended books about InfoVis: http://iv.homeunix.org/book.php

The most recommended book is:

Information Visualization
by: Robert Spence
This is the first fully integrated book on the emerging discipline of information visualization, incorporating dynamic examples on an accompanying website to complement the static representations within the book. Its emphasis is on real-world examples and applications of computer-generated/interactive information visualization. Information visualization deals with representing concepts and datain a meaningful way. Depending on the medium used, information can bevisualized in either static (e.g., a graph on a printed page) or dynamic forms. This book is appropriate for courses in information visualization, human-computer interaction, business information technology, and computer graphics.

Papers:

High Dimensional Data in Information Visualization, Sara Johansson

Useful Links:

Last updated 2012-02-10 by Tobias Åström (tobias.astrom@liu.se)