[
https://issues.apache.org/jira/browse/MAHOUT-1140?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
sam wu updated MAHOUT-1140:
---------------------------
Status: Patch Available (was: Open)
patch for RandomSeedGenerator.java using true uniform random sampling.
also attach TestUniformRandomSample.java for testing
> Uniform random sampling problem in RandomSeedGenerator.java
> -----------------------------------------------------------
>
> Key: MAHOUT-1140
> URL: https://issues.apache.org/jira/browse/MAHOUT-1140
> Project: Mahout
> Issue Type: New Feature
> Components: Clustering
> Affects Versions: 0.8
> Environment: all platform, generic
> Reporter: sam wu
> Priority: Minor
> Labels: features
> Fix For: 0.8
>
> Attachments: TestUniformRandomSample.java
>
> Original Estimate: 4h
> Remaining Estimate: 4h
>
> Uniform random sampling in RandomSeedGenerator.java is buggy. The selected
> samplings will be highly toward ending samples, eg: sample size=1000, k=3,
> almost all selected samples will be >=980th sample.
> This problem also exists in MiA code, ch09 at RandomPointUtils.java,
> chooseRandomPoints() method.
--
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
