tdmParaBootstrap

dset

name of column in dset which carries the target variable

resp

vector with names of input columns

input.variables

additional parameters [defaults in brackets]<dl class="dl-horizontal">
 <dt><code>ncopies</code></dt><dd>how many noisy copies to add</dd>
 <dt><code>ncsigma</code></dt><dd>[1.0] multiplicative factor for each std.dev.</dd>
 <dt><code>ncmethod</code></dt><dd>[3] which method to use for parametric bootstrap
 =1: each 'old' record from X in turn is the centroid for a new pattern;
 =2: the centroid is the average of all records from the same class, 
 the std.dev. is the same for all classes;
 =3: centroid as in '2', the std.dev. is the std.dev. of
 all records from the same class (*recommended*)</dd>
 <dt><code>TST.COL</code></dt><dd>(optional) name of column in dset where each PB record is marked with a 0</dd>
</dl>

opts

A normal distribution is approximated from the data given in <code>dset[,input.variables]</code> and new 
 data are drawn from this distribution for the columns <code>input.variables</code>. The column <code>resp</code> 
 is filled at random with levels with the same relative frequency as in <code>dset[,resp]</code>. 
 Other columns of dset are filled by copying the entries from the first row of dset.

Tuned Data Mining in R ('TDMR') performs the complete tuning of a
data mining task (predictive analytics, that is classification and regression).
Preprocessing parameters and modeling parameters can be tuned simultaneously.
It incorporates a variety of tuners (among them 'SPOT' and 'CMA' with package 'rCMA') and
allows integration of additional tuners. Noise handling in the data mining optimization
process is supported, see Koch et al. (2015) <doi:10.1016/j.asoc.2015.01.005>.

Wolfgang Konen

TDMR

Tuned Data Mining in R

tdmParaBootstrap function

additional parameters [defaults in brackets]<dl class='dl-horizontal'>
 <dt><code>ncopies</code></dt><dd>how many noisy copies to add</dd>
 <dt><code>ncsigma</code></dt><dd>[1.0] multiplicative factor for each std.dev.</dd>
 <dt><code>ncmethod</code></dt><dd>[3] which method to use for parametric bootstrap
 =1: each 'old' record from X in turn is the centroid for a new pattern;
 =2: the centroid is the average of all records from the same class, 
 the std.dev. is the same for all classes;
 =3: centroid as in '2', the std.dev. is the std.dev. of
 all records from the same class (*recommended*)</dd>
 <dt><code>TST.COL</code></dt><dd>(optional) name of column in dset where each PB record is marked with a 0</dd>
</dl>

tdmParaBootstrap: Parametric bootstrap: add 'noisy copies' to a data frame (training data).

Description

Usage

Arguments

Value

See Also