[jira] [Commented] (MAHOUT-516) Eigencuts produces unexpected results

["Martin Leginus (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MAHOUT-516?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13032374#comment-13032374 ] 

Martin Leginus commented on MAHOUT-516:
---------------------------------------

Hi guys,
I am trying to incorporate the Eigencuts clustering into my thesis and I found one problem within the code (maybe I am wrong :)). It seems for me that in EigencutsDriver.java on the line 134: "Matrix eigenVectors = new DenseMatrix(overshoot, eigenrank);" should be replaced with Matrix eigenVectors = new DenseMatrix(overshoot, dimensions); as the algorithm trims down the eigenvectors after DistributedLanczosSolver finishes eigenDecomposition. I believe that this part of the code should be identical with the SpectralKmeansDriver and therefore mentioned line should be replaced. I have used Mahout recently and I am just getting familiar with it, therefore its possible that my observation is wrong. If so I am sorry for spamming you. 
Regards 
Martin Leginus



> Eigencuts produces unexpected results
> -------------------------------------
>
>                 Key: MAHOUT-516
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-516
>             Project: Mahout
>          Issue Type: Bug
>          Components: Clustering
>    Affects Versions: 0.4
>            Reporter: Jeff Eastman
>            Assignee: Jeff Eastman
>             Fix For: 0.5
>
>         Attachments: MAHOUT-516.patch, jeastman.vcf
>
>
> Shannon reports he suspects a logic error in Eigencuts since it evidently does not produce exactly the expected results. It passes all current unit tests so we need to characterize the results differences and produce a test for it. Marking for 0.5 for now though we will fix it as soon as possible.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira