predict.outForest: Out-of-Sample Application

In outForest: Multivariate Outlier Detection and Replacement

Out-of-Sample Application

Description

Identifies outliers in new data based on previously fitted "outForest" object. The result of predict() is again an object of class "outForest". All its methods can be applied to it.

Usage

## S3 method for class 'outForest'
predict(
  object,
  newdata,
  replace = c("pmm", "predictions", "NA", "no"),
  pmm.k = 3L,
  threshold = object$threshold,
  max_n_outliers = Inf,
  max_prop_outliers = 1,
  seed = NULL,
  ...
)

Arguments

object	An object of class "outForest".
newdata	A new data.frame to be assessed for numeric outliers.
replace	Should outliers be replaced via predictive mean matching "pmm" (default), by "predictions", or by NA ("NA"). Use "no" to keep outliers as they are.
pmm.k	For replace = "pmm", from how many nearest OOB prediction neighbours (from the original non-outliers) to sample?
threshold	Threshold above which an outlier score is considered an outlier. The default is 3.
max_n_outliers	Maximal number of outliers to identify. Will be used in combination with threshold and max_prop_outliers.
max_prop_outliers	Maximal relative count of outliers. Will be used in combination with threshold and max_n_outliers.
seed	Integer random seed.
...	Further arguments passed from other methods.

Value

An object of class "outForest".

See Also

outForest(), outliers(), Data()

Examples

(out <- outForest(iris, allow_predictions = TRUE))
iris1 <- iris[1, ]
iris1$Sepal.Length <- -1
pred <- predict(out, newdata = iris1)
outliers(pred)
Data(pred)
plot(pred)
plot(pred, what = "scores")

outForest documentation built on May 31, 2023, 5:55 p.m.