Home
This Title All WIREs
WIREs RSS Feed
How to cite this WIREs title:
WIREs Comp Stat

Visualization toolkit software

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract Information visualization is a discipline that takes data as an input and produces visual representations of that data in a form that can be readily understood by people. The overall goal of a Visualization Toolkit is to provide a library that can be used by professionals to construct appropriate visualizations so as to allow their data to be understood and used in decision‐making. This article defines the major components of a visualization toolkit and provides details on how they can be implemented and the consequences and restrictions imposed by those decisions. WIREs Comput Stat 2012 doi: 10.1002/wics.1224 This article is categorized under: Algorithms and Computational Methods > Computer Graphics Statistical and Graphical Methods of Data Analysis > Statistical Graphics and Visualization

Display pipeline: Data to graphic element. This figure shows the steps in the processing chain that starts with the data and ends with the graphic element.

[ Normal View | Magnified View ]

A set of Consumer Price Index (CPI) data. These figures use four different indices; the Furniture CPI used in the previous two figures and three others. The left view shows a hierarchical clustering model of the data using a radial space‐filling layout. This view clusters months that are similar. The raw data for these variables have been plotted against time on the right as a line plot. A section of the clustering has been selected and shows in red (left) and as thick lines (right). It would be preferable to use the same aesthetic to display the selection in both views, but the difficulties of using the same aesthetic for lines and polygon elements forces compromises to be made.

[ Normal View | Magnified View ]

The same figure as figure 4, except with N = 6. Rather surprisingly for monthly data, a cycle of length 6 fits the data well—rather than the expected 12 month cycle, it appears that furniture prices work on a twice yearly cycle, at least for this data.

[ Normal View | Magnified View ]

A set of Consumer Price Index (CPI) data from the UK. In this figure we show the Furniture CPI. We have de‐trended the data using a loess smooth, and are exploring the cyclical nature using a very simple model which estimates a fixed set of divergences from the mean value for a cycle of length N. In this figure N = 4, which means we are fitting a seasonal model for months assuming a cycle of 5 months. Not surprisingly, the fit line (blue) does not match the de‐trended data (green) well.

[ Normal View | Magnified View ]

Human genome data. The horizontal axis shows a short sequence of 1000 nucleotides, color coded by type. Edges are created that link together runs within the genetic strand that have identical nucleotide patterns. Rather than display all such runs, only sequences of length at least 12 have been displayed. The color of the edges denotes the most common nucleotide in the matched sequence (using the mode statistic).

[ Normal View | Magnified View ]

US box‐office attendance for movies released in 2008. Each movie is represented by a density estimate over time, with the densities stacked on top of each other and wrapped around a circle to represent a year. The color used for each movie has no meaning; it is used simply to distinguish different movies. There is no edge correction in the density estimate, and so the densities around the beginning and end of the year cannot be taken as accurate.

[ Normal View | Magnified View ]

Browse by Topic

Statistical and Graphical Methods of Data Analysis > Statistical Graphics and Visualization
Algorithms and Computational Methods > Computer Graphics

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts