Func.SOD: Subspace outlier detection (SOD) algorithm

In HighDimOut: Outlier Detection Algorithms for High-Dimensional Data

Description

This function performs suspace outlier detection algorithm The implemented method is based on the work of Krigel, H.P., Kroger, P., Schubert, E., Zimek, A., Outlier detection in axis-parallel subspaces of high dimensional data, 2009.

Usage

Func.SOD(data, k.nn, k.sel, alpha = 0.8)

Arguments

data	is the data frame containing the observations. Each row represents an observation and each variable is stored in one column.
k.nn	specifies the value used for calculating the shared nearest neighbors. Note that k.nn should be greater than k.sel.
k.sel	specifies the number shared nearest neighbors. It can be interpreted as the number of reference set for constructing the subspace hyperplane.
alpha	specifies the lower limit for selecting subspace. 0.8 is set as default as suggested in the original paper.

Value

The function returns a vector containing the SOD outlier scores for each observation

Examples

library(ggplot2)
res.SOD <- Func.SOD(data = TestData[,1:2], k.nn = 10, k.sel = 5, alpha = 0.8)
data.temp <- TestData[,1:2]
data.temp$Ind <- NA
data.temp[order(res.SOD, decreasing = TRUE)[1:10],"Ind"] <- "Outlier"
data.temp[is.na(data.temp$Ind),"Ind"] <- "Inlier"
data.temp$Ind <- factor(data.temp$Ind)
ggplot(data = data.temp) + geom_point(aes(x = x, y = y, color=Ind, shape=Ind))

Example output

Warning message:
executing %dopar% sequentially: no parallel backend registered 

HighDimOut documentation built on May 2, 2019, 12:16 p.m.