[
https://issues.apache.org/jira/browse/MAHOUT-1167?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sebastian Schelter updated MAHOUT-1167:
---------------------------------------
Attachment: MAHOUT-1167.patch
Patch for the parallel precomputation. Also introduces an example for the
movielens1M dataset.
Will commit this in two days if there are no objections.
> Parallel item similarity precomputation on a single machine
> -----------------------------------------------------------
>
> Key: MAHOUT-1167
> URL: https://issues.apache.org/jira/browse/MAHOUT-1167
> Project: Mahout
> Issue Type: New Feature
> Components: Collaborative Filtering
> Affects Versions: 0.8
> Reporter: Sebastian Schelter
> Assignee: Sebastian Schelter
> Attachments: MAHOUT-1167.patch
>
>
> We need some code for item-based CF usecases with an intermediate data size
> (e.g., a few million interactions). In such cases, the data might be too big
> to allow online computation of similarities and recommendations, but at the
> same time, going to Hadoop might still not be necessary and desired.
> In such a case, it makes sense to precompute item similarities on a single
> machine.
--
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
