Decremental Reduction Optimization Procedures
Description
Similaritybased filters for removing label noise from a dataset as a preprocessing step of classification. For more information, see 'Details' and 'References' sections.
Usage
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17  ## S3 method for class 'formula'
DROP1(formula, data, ...)
## Default S3 method:
DROP1(x, k = 1, classColumn = ncol(x), ...)
## S3 method for class 'formula'
DROP2(formula, data, ...)
## Default S3 method:
DROP2(x, k = 1, classColumn = ncol(x), ...)
## S3 method for class 'formula'
DROP3(formula, data, ...)
## Default S3 method:
DROP3(x, k = 1, classColumn = ncol(x), ...)

Arguments
formula 
A formula describing the classification variable and the attributes to be used. 
data, x 
Data frame containing the tranining dataset to be filtered. 
... 
Optional parameters to be passed to other methods. 
k 
Number of nearest neighbors to be used. 
classColumn 
positive integer indicating the column which contains the (factor of) classes. By default, the last column is considered. 
Details
DROP1
goes over the dataset in the provided order, and removes those
instances whose removal does not decrease the accuracy of the 1NN rule in
the remaining dataset.
DROP2
introduces two modifications against DROP1
. Regarding the
order of processing instances, DROP2
starts with those which are
furthest from their nearest "enemy" (two instances are said to be "enemies"
if they belong to different classes). Moreover, DROP2
removes an
instance if its removal does not decrease the accuracy of the 1NN rule in
the original dataset (rather than the remaining dataset as in
DROP1
).
DROP3
is identical to DROP2
, but it includes a preprocessing
step to clean the borders between classes. It consists of applying the
ENN
method: any instance misclassified by its k nearest
neighbors is removed.
Value
An object of class filter
, which is a list with seven components:

cleanData
is a data frame containing the filtered dataset. 
remIdx
is a vector of integers indicating the indexes for removed instances (i.e. their row number with respect to the original data frame). 
repIdx
is a vector of integers indicating the indexes for repaired/relabelled instances (i.e. their row number with respect to the original data frame). 
repLab
is a factor containing the new labels for repaired instances. 
parameters
is a list containing the argument values. 
call
contains the original call to the filter. 
extraInf
is a character that includes additional interesting information not covered by previous items.
References
Wilson D. R., Martinez T. R. (2000): Reduction techniques for instancebased learning algorithms. Machine learning, 38(3), 257286. Wilson D. R., Martinez T. R. (1997, July): Instance pruning techniques. In ICML (Vol. 97, pp. 403411).
Examples
1 2 3 4 5 6 7 8 9  # Next example is not run in order to save time
## Not run:
data(iris)
trainData < iris[c(1:20,51:70,101:120),]
out1 < DROP1(Species~ Petal.Length + Petal.Width, data = trainData)
summary(out1, explicit = TRUE)
identical(out1$cleanData, trainData[setdiff(1:nrow(trainData),out1$remIdx),])
## End(Not run)
