# Package elki.clustering.subspace

Axis-parallel subspace clustering algorithms.

The clustering algorithms in this package are instances of both, projected clustering algorithms or subspace clustering algorithms according to the classical but somewhat obsolete classification schema of clustering algorithms for axis-parallel subspaces.

• Interface Summary
Interface Description
SubspaceClusteringAlgorithm<M extends SubspaceModel>
Interface for subspace clustering algorithms that use a model derived from SubspaceModel, that can then be post-processed for outlier detection.
• Class Summary
Class Description
CLIQUE
Implementation of the CLIQUE algorithm, a grid-based algorithm to identify dense clusters in subspaces of maximum dimensionality.
CLIQUE.Par
Parameterization class.
DiSH
Algorithm for detecting subspace hierarchies.
DiSH.DiSHClusterOrder
DiSH cluster order.
DiSH.Par
Parameterization class.
DOC
DOC is a sampling based subspace clustering algorithm.
DOC.Par
Parameterization class.
FastDOC
The heuristic variant of the DOC algorithm, FastDOC
FastDOC.Par
Parameterization class.
HiSC
Implementation of the HiSC algorithm, an algorithm for detecting hierarchies of subspace clusters.
HiSC.Par
Parameterization class.
P3C
P3C: A Robust Projected Clustering Algorithm.
P3C.ClusterCandidate
This class is used to represent potential clusters.
P3C.Par
Parameterization class.
P3C.Signature
P3C Cluster signature.
PreDeCon
PreDeCon computes clusters of subspace preference weighted connected points.
PreDeCon.Par
Parameterization class.
PreDeCon.Settings
Class containing all the PreDeCon settings.
PreDeCon.Settings.Par
Parameterization class.
PROCLUS
The PROCLUS algorithm, an algorithm to find subspace clusters in high dimensional spaces.
PROCLUS.DoubleIntInt
Simple triple.
PROCLUS.Par
Parameterization class.
PROCLUS.PROCLUSCluster
Encapsulates the attributes of a cluster.
SUBCLU<V extends NumberVector>
Implementation of the SUBCLU algorithm, an algorithm to detect arbitrarily shaped and positioned clusters in subspaces.
SUBCLU.Par<V extends NumberVector>
Parameterization class.
• Enum Summary
Enum Description
DiSH.Strategy
Available strategies for determination of the preference vector.