1 Talis, Birmingham, UK
Lecture at PROMISE Winter School 2012, Information Retrieval meets Information Visualization. Zinal, Valais - Switzerland 23 - 27 January 2012
Download: draft chapter (PDF, 1Mb); slides as PDF (2 up, 8.4M), PPT (8.6M)
Now part of Springer volume: A. Dix (2013). Introduction to Information Visualisation. Information Retrieval Meets Information Visualization, LNCS 7757, 2013, pp 1-27
This is a short introduction to information visualisation, which is essentially about making data easier to understand using direct sensory experience, as opposed to more linguistic/logical reasoning.
Note that while the term suggests the visual sense, and indeed this is the most common, in fact there are also aural and tactile ‘visualisations', for use by those with disabilities or in situations where visual displays would not be appropriate. Visualisations may be designed for end users / information consumers for whom the visualisation will normally have to be very easy to understand, or for more professional data analysts, who may be able to use more complex visualisations including explicit training. Most challenging is exploratory visualisation as, by definituion, you do not know before hand what you want to be absle to see in the data.
Static visualisation can be traced back at least 2500 years, including linegraphs from the 10th Century, but it was in the early 1990s that computer-based interactive visualisations became possible. Currently, visualisation needs to address data sets of rapidly growing sizes, but also is often presented over the web, both creating new challenges. Furthermore, the increasing complexity of data processing, including machine learning, that is part of visualisation has led to the new arae of visual analytics.
It is easy to focus on the single user interacting witgh a visual represention (visualisation) or with it and the data processing leaidng to it (visual analytics). However, both take place in a broader context where the data being visualised comes from the world, the visualisation helps indioviduals or organisations to make decsions, and these decisions lead to actions which change the world and thus the data being visualised.
Designing visualisations requires an understaidng of the human visual (or other sense) system, the obhjective/goals that the visualisation is to aid, and aesthetics (especially for persuasive graphics). Choosing an appropriate represnetaton typically requires choices and trade-offs between factors, but interactive visualisation can soften these trade-offs, allowing choices to be remade by the user as they interact with the system.
References and Links
Books on information visualisation
Readings in Information Visualization
Mastering the Information Age
Illuminating the Path
Some web sites collecting visualisations
Guardian Data: Deficit Debt Government Borrowing Data
IBM Many Eyes
Information is Beautiful
Specific systems and papers
Map Your Move
Interactive stacked histogram.
Alan Dix 27/1/2012