[jira] Updated: (MAHOUT-157) Frequent Pattern Mining using Parallel FP-Growth

["Robin Anil (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MAHOUT-157?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Robin Anil updated MAHOUT-157:
------------------------------

    Attachment: MAHOUT-157-September-5.patch

Improvements in Code structure.
Massive performance improvement 10x improvement in speed. Takes half the memory.
50K transactions with a minSupport 10K takes 15 second instead of 2:20 min in the earlier patch.(thanks to YourKit profiling and multiple iterations of 80/20 rule).  

Current bottle necks(could lead to at max 30% reduction in runtime)
* Need for a faster HashMap. (there are a lot of put/get) Could use some primitive representation there.
* Need a faster implementation of a bounded SortedSet. Currently use an encapsulated TreeSet where least value is removed after every insert call  if the max capacity is found to be exceeded. Another frequent operation is: merge two bounded SortedSets 

Updated Tests


> Frequent Pattern Mining using Parallel FP-Growth
> ------------------------------------------------
>
>                 Key: MAHOUT-157
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-157
>             Project: Mahout
>          Issue Type: New Feature
>    Affects Versions: 0.2
>            Reporter: Robin Anil
>             Fix For: 0.2
>
>         Attachments: MAHOUT-157-August-17.patch, MAHOUT-157-August-24.patch, MAHOUT-157-August-31.patch, MAHOUT-157-August-6.patch, MAHOUT-157-Combinations-BSD-License.patch, MAHOUT-157-Combinations-BSD-License.patch, MAHOUT-157-inProgress-August-5.patch, MAHOUT-157-September-5.patch
>
>
> Implement: http://infolab.stanford.edu/~echang/recsys08-69.pdf

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.