Constructs a parallel coordinate plot for a data set with classes in last column.
Usage
parallelplot(x, name = "", comb = -1, class = 0, obs = rep(0, 0), col = 2, lty = 1, ...)
Arguments
x
A matrix of numerical values with classes in last column
name
The name of data set as will appear in the graph title
comb
An integer that represents the number of one of the possible
combinations for the columns of this matrix.
class
A value representing the class number to which the plot should be limited
obs
A list of one or more row numbers that are to be highlighted in the plot
col
A value that provides a choice of color for the plot (if plotting only one class)
lty
A value that provides a choice of line width for the plot (if plotting only one class)
...
Additional arguments for the matplot function
Value
A parallel coordinate plot of the data is produced.
Details
This plot is not recommended for a large number of features (say more than 50).
If comb=0, all distinct combinations of columns are graphed. If comb=-1 (default), the attributes are
plotted in their original order, else comb should be equal to an integer that represents
the number of one of the possible combinations for the columns of this matrix.
References
Wegman, E. (1990), Hyperdimensional data analysis using parallel coordinates,
Journal of the American Statistical Association,85,664-675