kendall.plot: Plot of the Kendall Criterion values

Description Usage Arguments Details Value Author(s) See Also Examples

Description

Plot of the Kendall Criterion values of aggregate ranked lists; useful for comparing performances of several algorithms.

Usage

1
2
Kendall.plot(input, all.aggregates, space = NULL, algorithm = NULL, p =
0.5, w = NULL, ...)

Arguments

input

A list object containing individual ranked lists.

all.aggregates

A list comprising of aggregate top-k lists from different algorithms to be compared.

space

A list containing the underlying spaces. If not explicitly specified, all lists are treated as coming from a common space defined by the union of all input lists.

algorithm

A vector listing the names corresponding to the algorithms used to construct the aggregate ranked lists all.aggregates.

p

A parameter between 0 and 1 for setting the distance of a pair of elements between two lists, if at least one of the elements is not in the underlying space of one of the list or if both elements belong to one list but neither belong to the other list. (We recommend using p=0.5 for a "neutral approach".)

w

Weight vector assigned to the input list. Prior information on the reliability of each input list can be incorporated. The default is set to equal weight for each input list.

...

Other parameters passed on to the plot function.

Details

Compute the weighted Kendall's distance between each of the aggregate ranked list with the input ranked lists and plot the computed distances.

Value

A plot of Kendall's distance for each of the aggregate list.

Author(s)

Shili Lin <shili@stat.osu.edu

See Also

KendallMLists

Examples

1
2
3
4
5
#get sample data
data(TopKSpaceSampleInput)
outMC=MC(input,space)
all.aggregate=list(outMC$MC1.TopK,outMC$MC2.TopK,outMC$MC3.TopK)
Kendall.plot(input, all.aggregate,space, algorithm=c("MC1","MC2","MC3"))


Search within the TopKLists package
Search all R packages, documentation and source code

Questions? Problems? Suggestions? or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.