Synopsis

A filter that applies the LOF (Local Outlier Factor) algorithm to compute an "outlier" score for each instance in the data. Can use multiple cores/cpus to speed up the LOF computation for large datasets. Nearest neighbor search methods and distance functions are pluggable.

Available via the package management system for Weka >= 3.7.2 (localOutlierFactor)

For more information, see:

Markus M. Breunig, Hans-Peter Kriegel, Raymond T. Ng, Jorg Sander (2000). LOF: Identifying Density-Based Local Outliers. ACM SIGMOD Record. 29(2):93-104.

Options

The table below describes the options available for LOF.

Option	Description
NNSearch	The nearest neighbour search algorithm to use (Default: weka.core.neighboursearch.LinearNNSearch).
minPointsLowerBound	The lower bound (minPtsLB) to use on the range for k when determining the maximum LOF value
minPointsUpperBound	The upper bound (minPtsUB) to use on the range for k when determining the maximum LOF value
numExecutionSlots	The number of execution slots (threads) to use for finding LOF values.

Capabilities

The table below describes the capabilities of LOF.

Capability	Supported
Class	Binary class, No class, Nominal class, Missing class values, Numeric class, Date class
Attributes	Numeric attributes, Binary attributes, Nominal attributes, Empty nominal attributes, Missing values, Unary attributes, Date attributes
Min # of instances	0

Browser not supported