LOOP {DDoutlier}

Local Outlier Probability (LOOP) algorithm

Description

Function to calculate the Local Outlier Probability (LOOP) as an outlier score for observations. Suggested by Kriegel, H.-P., Kröger, P., Schubert, E., & Zimek, A. (2009)

Usage

LOOP(dataset, k = 5, lambda = 3)

Arguments

dataset

The dataset for which observations have a LOOP score returned

k

The number of k-nearest neighbors to compare density with

lambda

Multiplication factor for standard deviation. The greater lambda, the smoother results. Default is 3 as used in original papers experiments

Details

LOOP computes a local density based on probabilistic set distance for observations, with a user-given k-nearest neighbors. The density is compared to the density of the respective nearest neighbors, resulting in the local outlier probability. The values ranges from 0 to 1, with 1 being the greatest outlierness. A kd-tree is used for kNN computation, using the kNN() function from the 'dbscan' package. The LOOP function is useful for outlier detection in clustering and other multidimensional domains

Value

A vector of LOOP scores for observations. 1 indicates outlierness and 0 indicate inlierness

Author(s)

Jacob H. Madsen

References

Kriegel, H.-P., Kröger, P., Schubert, E., & Zimek, A. (2009). LoOP: Local Outlier Probabilities. In ACM Conference on Information and Knowledge Management, CIKM 2009, Hong Kong, China. pp. 1649-1652. DOI: 10.1145/1645953.1646195

Examples

# Create dataset
X <- iris[,1:4]

# Find outliers by setting an optional k
outlier_score <- LOOP(dataset=X, k=10, lambda=3)

# Sort and find index for most outlying observations
names(outlier_score) <- 1:nrow(X)
sort(outlier_score, decreasing = TRUE)

# Inspect the distribution of outlier scores
hist(outlier_score)