[
https://issues.apache.org/jira/browse/MAHOUT-362?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12853311#action_12853311
]
Jake Mannix commented on MAHOUT-362:
------------------------------------
It would be really nice if this code could be extend outside of the realm of
Taste. This is a very general sparse matrix operation, and should probably live
in the core o.a.m.math.hadoop package, operating on general sparse vectors.
I can adapt it to that package after it's committed though, no need to hold it
up on account of "insufficient generality" :)
> Computation of pairwise cosine similarities for Item-Based Collaborative
> Filtering
> ----------------------------------------------------------------------------------
>
> Key: MAHOUT-362
> URL: https://issues.apache.org/jira/browse/MAHOUT-362
> Project: Mahout
> Issue Type: New Feature
> Components: Collaborative Filtering
> Reporter: Sebastian Schelter
> Attachments: MAHOUT-362.patch
>
>
> Provides a map/reduce job to precompute the pairwise cosine similarities
> between the item vectors of the user-item-matrix.
> The code uses a slightly modified version of the algorithm suggested in
> "Elsayed et al: Pairwise Document Similarity in Large Collections with
> MapReduce"
> (http://www.umiacs.umd.edu/~jimmylin/publications/Elsayed_etal_ACL2008_short.pdf)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
