lofactor: An implementation of the LOF algorithm

In DMwR2: Functions and Data for the Second Edition of "Data Mining with R"

Description

This function obtain local outlier factors using the LOF algorithm. Namely, given a data set it produces a vector of local outlier factors for each case.

Usage

lofactor(data, k)

Arguments

data	A data set that will be internally coerced into a matrix.
k	The number of neighbours that will be used in the calculation of the local outlier factors.

Details

This function re-implements the code previously made available in the dprep package (Acuna et. al., 2009) that was removed from CRAN. This code in turn is an implementation of the LOF method by Breunig et. al. (2000). See this reference to understand the full details on how these local outlier factors are calculated for each case in a data set.

Value

The function returns a vector of local outlier factors (numbers). This vector has as many values as there are rows in the original data set.

Author(s)

Luis Torgo [e-mail]

References

Acuna, E., and Members of the CASTLE group at UPR-Mayaguez, (2009). dprep: Data preprocessing and visualization functions for classification. R package version 2.1.

Breunig, M., Kriegel, H., Ng, R., and Sander, J. (2000). LOF: identifying density-based local outliers. In ACM Int. Conf. on Management of Data, pages 93-104.

Torgo, L. (2016) Data Mining using R: learning with case studies, second edition, Chapman & Hall/CRC (ISBN-13: 978-1482234893).

http://ltorgo.github.io/DMwR2

Examples

data(iris)
lof.scores <- lofactor(iris[,-5],10)

Example output

DMwR2 documentation built on May 2, 2019, 1:42 p.m.