Class OPTICS<T>
java.lang.Object
kieker.analysis.generic.clustering.optics.OPTICS<T>
- Type Parameters:
T
- model type
public class OPTICS<T>
extends java.lang.Object
An implementation of the OPTICS algorithm. A detailed explanation of the algorithm can be found
in the paper "OPTICS: ordering points to identify the clustering structure".
- Since:
- 2.0.0
- Author:
- Lars Jürgensen
-
Constructor Summary
Constructors Constructor Description OPTICS(MTree<OpticsData<T>> mtree, double maxDistance, int minPTs, java.util.List<OpticsData<T>> models)
-
Method Summary
Modifier and Type Method Description java.util.List<OpticsData<T>>
calculate()
This calculates the OPTICS result.double
getMaxDistance()
int
getMinPTs()
MTree<OpticsData<T>>
getMtree()
-
Constructor Details
-
OPTICS
public OPTICS(MTree<OpticsData<T>> mtree, double maxDistance, int minPTs, java.util.List<OpticsData<T>> models)- Parameters:
mtree
- The M-Tree with the behavior models to be clusteredmaxDistance
- The maximal distance two neighbors can have (the epsilon value of the algorithm)minPTs
- The minimal amount of neighbors a object must have to be called a core-objectmodels
- A list of all models to be clustered. They have to be the same as the models in the M-Tree
-
-
Method Details
-
calculate
This calculates the OPTICS result.- Returns:
- An ordered list of the behavior models. The reachability distances of the models are important for the evaluation.
-
getMinPTs
public int getMinPTs() -
getMaxDistance
public double getMaxDistance() -
getMtree
-