Description Usage Arguments Details
Creates a group of names with n members from two groups, such that all names are as similar to each other as possible. Names are not only similar across groups but also to each other in the group.
1 2 | match.groups(split, n, discard = 0, subset = filter.names(),
ga.params = list(), ...)
|
split |
The name of the rating on which the split should be performed |
n |
Number of names from each group to select |
discard |
The percentage of names too close to the median which should be discarded (default: 0, i.e. keep all names) |
subset |
An optional subset on which the split should be done. If this is left out, the split will be created on all names. |
ga.params |
Parameters to genetic algorithm
|
... |
Arguments passed on to
|
Uses a genetic algorithm to find the best names, so different results may be found each run.
To seed the genetic algorithm with well suited sets, first random sets of names are created. To increase the search space, more names are selected than will later be retrieved for each of the partitions. The number of additional names, which are included is controlled by a ga.param value.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.