[ https://issues.apache.org/jira/browse/MAHOUT-887?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150567#comment-13150567 ]
Jeff Eastman edited comment on MAHOUT-887 at 11/15/11 5:03 PM: --------------------------------------------------------------- Take a look at MeanShiftCanopy. It implements a bottom-up, iterative, agglomerative clustering algorithm based upon Mean Shift clustering. In it, each input vector is converted to a canopy and T1/T2 distances are used to merge adjacent canopies in each iteration. The centroid computation done at the end of each iteration shifts the centers of the remaining canopies to their mean positions, hence the name. The MSC implementation has some scalability issues which are common to agglomerative algorithms. I've made a couple of recent changes to improve things but they have not been tested much AFAICT. You could look into some of these issues if you wanted. was (Author: jeastman): Take a look at MeanShiftCanopy. It implements a bottom-up, iterative, agglomerative clustering algorithm based upon Mean Shift clustering. In it, each input vector is converted to a canopy and T1/T2 distances are used to merge adjacent canopies in each iteration. The centroid computation done at the end of each iteration shifts the centers of the remaining canopies to their mean positions, hence the name. > Bottom Up Clustering > -------------------- > > Key: MAHOUT-887 > URL: https://issues.apache.org/jira/browse/MAHOUT-887 > Project: Mahout > Issue Type: New Feature > Components: Clustering > Affects Versions: 0.6 > Environment: Linux Windows > Reporter: Paritosh Ranjan > Labels: features > Fix For: 0.6 > > > Bottom up clustering is achieved by starting with small clusters/single > points and then merging clusters recursively which are closer than a > specified control constraint. -- 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