[
https://issues.apache.org/jira/browse/SPARK-17459?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16343051#comment-16343051
]
sandeep commented on SPARK-17459:
---------------------------------
is there an update on this ? or did spark implement an alternative that is
recommended ?
I know that PCA exists, but LDA is much better for two class separation
> Add Linear Discriminant to dimensionality reduction algorithms
> --------------------------------------------------------------
>
> Key: SPARK-17459
> URL: https://issues.apache.org/jira/browse/SPARK-17459
> Project: Spark
> Issue Type: New Feature
> Components: ML
> Reporter: Joshua Howard
> Priority: Minor
>
> The goal is to add linear discriminant analysis as a method of dimensionality
> reduction. The algorithm and code are very similar to PCA, but instead
> project the data set onto vectors that provide class separation. LDA is a
> more effective alternative to PCA in terms of preprocessing for
> classification algorithms.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
