Information Visualisation consists of two steps:

Analysis: data, especially when the quantity is large, needs to be analysed before it is visualised. This step can be performed by:

• data mining (most suitable for financial data)
• parsing (for software reengineering)
• human analysis (for crime data), or
• a combination of these three.

The analysis essentially reduces the size of the original data either by forming an abstraction, or summary, or focussing on a small part of the data

The first method gives an overview but omits details, the second gives a detailed view but the user can get lost with no overview. One of the challenges of visualisation is to provide an abstraction of the whole data set as well as a detailed view of part of it.

Creating a picture: the creative step in visualisation is to make a picture of the reduced data. This involves designing a visual metaphor for the data, more of an art than a science, and creating an appropriate layout, especially where the data is highly abstract and has no preconceived geometry. These two processes result in a readable picture. To improve the use of the picture we create navigation tools allowing you to move seamlessly from one view to another, and we employ the best available computer graphics such as data walls and immersive displays.

In recent years, information visualisation has faced two challenges:

• The size of the data to be visualised; this has grown out of proportion to processor speed, storage capacity, and (most relevant to visualisation) screen size.
• The speed of data generation: in the past, one can assume that the data is relatively static. Recent advances in networks have streamed data at a high rate, and visualisation systems must react quickly to changes.

The data held by CMCRC has both these properties. The visualisation challenge continues to be an exciting and demanding pursuit of the CMCRC.