You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@spark.apache.org by "sandeep (JIRA)" <ji...@apache.org> on 2018/01/29 08:20:00 UTC

[jira] [Commented] (SPARK-17459) Add Linear Discriminant to dimensionality reduction algorithms

    [ 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-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org