[
https://issues.apache.org/jira/browse/MAHOUT-1066?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13481441#comment-13481441
]
Matt Molek commented on MAHOUT-1066:
------------------------------------
There is an old feature request for this. It looks like there was some progress
on it, but I'm not sure what ever became of it:
https://issues.apache.org/jira/browse/MAHOUT-521
> How to generate sparsed Vectors from the specified dictionary.
> --------------------------------------------------------------
>
> Key: MAHOUT-1066
> URL: https://issues.apache.org/jira/browse/MAHOUT-1066
> Project: Mahout
> Issue Type: Question
> Components: Clustering
> Affects Versions: 0.7
> Reporter: Hiroaki Kubota
> Assignee: Paritosh Ranjan
> Fix For: 0.8
>
>
> I'd like to do clustering our natural language data.
> The first, I used the 'seq2sparse' command to vectorize our data.
> I got sparsed vectors and a dictionary.
> And we could do k-means and got suitable clusters.
> It was OK.
> The next, I'd like to add some data to previous calculated clusters.
> So I want to get additional vectors from new additional data based on
> previous dictionary.
> Probably I think,
> It is impossible to get really accurate vectors by using only additional data.
> However,I'd like to reduce processing time so It's OK if I get the vector
> that is useful for decision tree.
> Please give me advice !
> Regard,
--
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
