[
https://issues.apache.org/jira/browse/MAHOUT-15?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jeff Eastman updated MAHOUT-15:
-------------------------------
Attachment: MAHOUT-15b.patch
I noticed that the mergeCanopy method was only adding the new canopy center to
the existing canopies, but not adding the existing canopy centers to itself.
This produced an assymetry defect that, I think, also exists in canopy
clustering and kmeans. Correcting this problem leads the algorithm to now
converge exactly upon the input image after 40-odd iterations:
ABBBBBBBBB
BABBBBBBBB
BBABBBBBBB
BBBABBBBBB
BBBBABBBBB
BBBBBABBBB
BBBBBBABBB
BBBBBBBABB
BBBBBBBBAB
BBBBBBBBBA
> Investigate Mean Shift Clustering
> ---------------------------------
>
> Key: MAHOUT-15
> URL: https://issues.apache.org/jira/browse/MAHOUT-15
> Project: Mahout
> Issue Type: New Feature
> Components: Clustering
> Reporter: Jeff Eastman
> Assignee: Jeff Eastman
> Attachments: MAHOUT-15a.patch, MAHOUT-15b.patch
>
>
> "The mean shift algorithm is a nonparametric clustering technique which does
> not require prior knowledge of the number of clusters, and does not constrain
> the shape of the clusters."
> http://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/TUZEL1/MeanShift.pdf
> Investigate implementing mean shift clustering using Hadoop
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
