You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Apache Spark (JIRA)" <ji...@apache.org> on 2016/06/14 07:24:40 UTC
[jira] [Commented] (SPARK-15938) Adding "support" property to MLlib
Association Rule
[ https://issues.apache.org/jira/browse/SPARK-15938?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15329009#comment-15329009 ]
Apache Spark commented on SPARK-15938:
--------------------------------------
User 'hhbyyh' has created a pull request for this issue:
https://github.com/apache/spark/pull/13656
> Adding "support" property to MLlib Association Rule
> ---------------------------------------------------
>
> Key: SPARK-15938
> URL: https://issues.apache.org/jira/browse/SPARK-15938
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Reporter: yuhao yang
> Priority: Minor
>
> _Support_ is an indication of how frequently the item-set appears in the database. Besides confidence, "Support" is another critical property for Association rule.
> References:
> https://en.wikipedia.org/wiki/Association_rule_learning
> http://www.philippe-fournier-viger.com/spmf/index.php?link=documentation.php#allassociationrules
> https://www-users.cs.umn.edu/~kumar/dmbook/ch6.pdf
> _Support_ can be either the count of appearances or the fraction within the dataset. I choose to use the count as:
> 1. API compatibility: Currently both FPGrowthModel and Association Rule does not have the information about size of the dataset. I'd try to avoid breaking a list of public APIs.
> 2. This also refers to the API of SPMF. http://www.philippe-fournier-viger.com/spmf/index.php?link=documentation.php#allassociationrules.
> In the next steps, we could add constraint like minSupport as in other libraries. FPGrowthModel should also contains the size of the dataset.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org