Rlof (version 1.1.1)

lof: Local Outlier Factor

Description

A function that finds the local outlier factor (Breunig et al.,2000) of the matrix "data" using k neighbours. The local outlier factor (LOF) is a measure of outlierness that is calculated for each observation. The user decides whether or not an observation will be considered an outlier based on this measure. The LOF takes into consideration the density of the neighbourhood around the observation to determine its outlierness. This is a faster implementation of LOF by using a different data structure and distance calculation function compared to lofactor() function available in dprep package. It also supports multiple k values to be calculated in parallel, as well as various distance measures besides the default Euclidean distance.

Usage

lof(data, k, cores = NULL, ...)

Arguments

data

The data set to be explored, which can be a data.frame or matrix

k

The kth-distance to be used to calculate LOFs. k can be a vector which contains multiple k values based on which LOFs need to be calculated.

cores

optional, The number of cores to be used for parallel computing. If not provided, the maximum number of cores available is used by default.

...

The parameters to be passed to distmc() function, specifying the distance measure.

Value

lof

A matrix with the local outlier factor of each observation as rows and each k value as columns

Details

The LOFs are calculated over multiple k values in parallel, and the maximum number of the cpus will be utilised to achieve the best performance.

References

Breuning, M., Kriegel, H., Ng, R.T, and Sander. J. (2000). LOF: Identifying density-based local outliers. In Proceedings of the ACM SIGMOD International Conference on Management of Data.

Examples

# NOT RUN {
---- Detecting the top outliers using the LOF algorithm
# }
# NOT RUN {
---- with k = 5,6,7,8,9 and 10, respectively----
# }
# NOT RUN {
data(iris)
df<-iris[-5]
# }
# NOT RUN {
df.lof<-lof(df,c(5:10),cores=2)
# }