@Title(value="OutRank: ranking outliers in high dimensional data") @Description(value="Ranking outliers in high dimensional data - score 1") @Reference(authors="Emmanuel M\u00fcller, Ira Assent, Uwe Steinhausen, Thomas Seidl", title="OutRank: ranking outliers in high dimensional data", booktitle="Proc. 24th Int. Conf. on Data Engineering (ICDE) Workshop on Ranking in Databases (DBRank)", url="https://doi.org/10.1109/ICDEW.2008.4498387", bibkey="DBLP:conf/icde/MullerASS08") public class OutRankS1 extends AbstractAlgorithm<OutlierResult> implements OutlierAlgorithm
Algorithm to score outliers based on a subspace clustering result. This class implements score 1 of the OutRank publication, which is a score based on cluster sizes and cluster dimensionality.
Reference:
E. Müller, I. Assent, U. Steinhausen, T. Seidl
OutRank: ranking outliers in high dimensional data
In Proceedings 24th International Conference on Data Engineering (ICDE)
Workshop on Ranking in Databases (DBRank)
Modifier and Type | Class and Description |
---|---|
static class |
OutRankS1.Parameterizer
Parameterization class.
|
Modifier and Type | Field and Description |
---|---|
(package private) double |
alpha
Weighting parameter of size vs. dimensionality score.
|
protected SubspaceClusteringAlgorithm<? extends SubspaceModel> |
clusteralg
Clustering algorithm to run.
|
private static Logging |
LOG
The logger for this class.
|
ALGORITHM_ID
Constructor and Description |
---|
OutRankS1(SubspaceClusteringAlgorithm<? extends SubspaceModel> clusteralg,
double alpha)
Constructor.
|
Modifier and Type | Method and Description |
---|---|
TypeInformation[] |
getInputTypeRestriction()
Get the input type restriction used for negotiating the data query.
|
protected Logging |
getLogger()
Get the (STATIC) logger for this class.
|
OutlierResult |
run(Database database)
Runs the algorithm.
|
private static final Logging LOG
protected SubspaceClusteringAlgorithm<? extends SubspaceModel> clusteralg
double alpha
public OutRankS1(SubspaceClusteringAlgorithm<? extends SubspaceModel> clusteralg, double alpha)
clusteralg
- SubspaceClusteringAlgorithm
to usealpha
- Alpha parameter to balance size and dimensionality.public OutlierResult run(Database database)
Algorithm
run
in interface Algorithm
run
in interface OutlierAlgorithm
run
in class AbstractAlgorithm<OutlierResult>
database
- the database to run the algorithm onpublic TypeInformation[] getInputTypeRestriction()
AbstractAlgorithm
getInputTypeRestriction
in interface Algorithm
getInputTypeRestriction
in class AbstractAlgorithm<OutlierResult>
protected Logging getLogger()
AbstractAlgorithm
getLogger
in class AbstractAlgorithm<OutlierResult>
Copyright © 2019 ELKI Development Team. License information.