find_spatial_outlier: Generate two dimensional feature space from tensor data.

In pridiltal/mask: Anomaly Detection in Multivariate Spatio-temporal data with K-measurements

Description

This function converts the original three way dataset into two dimensional feature space.

Usage

find_spatial_outlier(
  data,
  ncomp = 2,
  center = TRUE,
  center.mode = c("A", "B", "C"),
  scale = TRUE,
  scale.mode = c("A", "B", "C"),
  alpha = 0.01,
  tensor_decom = c("Tucker3", "parafac")
)

Arguments

data	3-way array of data, A mode- space
ncomp	Number of components. Default is set to 2
center	Whether to center the data
center.mode	If centering the data, on which mode to do this
scale	Whether to scale the data
scale.mode	scaling the data, on which mode to do this
alpha	Threshold for determining the cutoff for outliers.
tensor_decom	Method to decompose the 3-way array

Value

a list with components

outliers	The indexes of the spatial observations determined to be outliers.
out_scores	Outliying score of each spatial observation
type	Type of each spatial observation: outlier or typical
plot_data	A-mode component data with ourlying measures

Examples

library(rrcov3way)
data(Arno)
result <- find_spatial_outlier(
  data = Arno, tensor_decom = "Tucker3"
)
result$out_data

pridiltal/mask documentation built on Dec. 22, 2021, 9:51 a.m.