[
https://issues.apache.org/jira/browse/MATH-917?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Thomas Neidhart updated MATH-917:
---------------------------------
Attachment: clustering.zip
I have attached the result of some refactoring of the cluster package.
It includes the following changes/additions:
* Move distance calculation from the Clusterable interface to a dedicated
DistanceMeasure interface with a first concrete implementation:
EuclideanDistance
* Modify a Cluster to make it more general: remove center as this is only used
for centroid based clustering algos
* Introduce a Clusterer interface with currently only one method:
cluster(Collection<Clusterable>), but we may add more, e.g. with a
maxIterations argument
* Added an AbstractClusterer class which provides basic stuff for each
clusterer, e.g. a distance measure.
* The existing clustering algos implement the new interface via the abstract
class
I like the Clusterable interface, as it makes it quite easy to extend existing
data objects to make them an input for the clusterer. The simple *Point
implementations have been kept but I am not fully happy with the name.
I would like to get feedback if this goes in the right direction, and if so,
will finish the contribution.
> More distance measurements are needed in o.a.c.m.stat.clustering.
> -----------------------------------------------------------------
>
> Key: MATH-917
> URL: https://issues.apache.org/jira/browse/MATH-917
> Project: Commons Math
> Issue Type: Improvement
> Reporter: Reid Hochstedler
> Fix For: 4.0
>
> Attachments: clustering.zip
>
>
> Currently only Euclidean distance is used for distance measurement, it would
> be easy to quickly add Manhattan and Chebyshev distance among others.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
