Togaware DATA MINING
Desktop Survival Guide
by Graham Williams
Google

Brushing

Often we will identify groups within a plot through manual brushing. We also want to obtain the data rows that correspond to the brushed points. GGobi has two concepts: a selection is the collection of points which fall within the current brush whilst a shadow is the collection of points that have been selected over time.

To obtain the list of selected or shadowed points we can use the selected or the shadowed functions from rggobi. These are applied to GGobiData objects. We can access such an object by saving the object returned when we start up GGobi:



> library(rattle)
> data(audit)
> gg <- ggobi(audit)
# Choose, for example, Age for X and Income for Y
# Select Brush from Interaction
# Select points by moving yellow brush with left mouse button
# Middle mouse button to change size of the brush
> which(selected(gg[1]))
[1]  314  547  558  596  921 1512

A selection consists of just those points currently within the brush. To identify a collection of points of interest we can shadow the points. Choose Shadow from Point brushing and turn on Persistent. Then brush points of interest, to create shadows. We can then extract the full set of shadowed points.



> which(shadowed(gg[1]))
 [1]   42   79   96  124  206  289  319  552  578  586  601  727  
[13]  792  873  956 1040 1043 1044 1054 1090 1132 1310 1348 1351
[25] 1377 1501 1553 1567 1573 1586 1629 1666 1794 1878 1963 1983



Copyright © 2004-2010 Togaware Pty Ltd
Support further development through the purchase of the PDF version of the book.
The PDF version is a formatted comprehensive draft book (with over 800 pages).
Brought to you by Togaware. This page generated: Sunday, 22 August 2010