[
https://issues.apache.org/jira/browse/SPARK-8871?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15390841#comment-15390841
]
Jonathan Svirsky commented on SPARK-8871:
-----------------------------------------
Hi Sean, You are right. The main idea under the proposed improvements (closed
and maximal sets) is to reduce the output of the fpgrowth. It is also more
useful to see the maximal frequent itemset X (all its subsets together) instead
of a set of frequent itemsets which compose this superset X.
> Add maximal frequent itemsets filter in Spark MLib FPGrowth
> -----------------------------------------------------------
>
> Key: SPARK-8871
> URL: https://issues.apache.org/jira/browse/SPARK-8871
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Reporter: Jonathan Svirsky
>
> Maximal frequent itemsets can be exctracted as all root-to-leaf paths(sets)
> from FP-Trees.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
