You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by "Paritosh Ranjan (JIRA)" <ji...@apache.org> on 2012/08/13 04:13:38 UTC

[jira] [Created] (MAHOUT-1054) Use ball KMeans for clustering

Paritosh Ranjan created MAHOUT-1054:
---------------------------------------

             Summary: Use ball KMeans for clustering
                 Key: MAHOUT-1054
                 URL: https://issues.apache.org/jira/browse/MAHOUT-1054
             Project: Mahout
          Issue Type: Improvement
          Components: Clustering
            Reporter: Paritosh Ranjan
             Fix For: 0.8


Use ball KMeans for clustering.

Ted has already implemented a similar version. http://github.com/tdunning/knn

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

[jira] [Commented] (MAHOUT-1054) Use ball KMeans for clustering

Posted by "Ted Dunning (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/MAHOUT-1054?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13432896#comment-13432896 ] 

Ted Dunning commented on MAHOUT-1054:
-------------------------------------

See https://github.com/tdunning/knn/blob/master/src/main/java/org/apache/mahout/knn/cluster/BallKmeans.java

for reference.  I intend to bring this in during the 0.8 process.

Pulling the code in early is fine by me, but the current knn stuff has some serious deficiencies in that it doesn't yet implement the Mahout standard for clustering algorithms, nor does it have a map-reduce implementation.  The ball k-means is also currently untested.
                
> Use ball KMeans for clustering
> ------------------------------
>
>                 Key: MAHOUT-1054
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-1054
>             Project: Mahout
>          Issue Type: Improvement
>          Components: Clustering
>            Reporter: Paritosh Ranjan
>             Fix For: 0.8
>
>
> Use ball KMeans for clustering.
> Ted has already implemented a similar version. http://github.com/tdunning/knn

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira