spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Joseph K. Bradley (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SPARK-18334) MinHash should use binary hash distance
Date Fri, 18 Nov 2016 20:31:58 GMT

    [ https://issues.apache.org/jira/browse/SPARK-18334?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15677667#comment-15677667
] 

Joseph K. Bradley commented on SPARK-18334:
-------------------------------------------

Adding a note: Per discussions on PRs, we need to discuss this more and also think about it
in terms of [SPARK-18454].

> MinHash should use binary hash distance
> ---------------------------------------
>
>                 Key: SPARK-18334
>                 URL: https://issues.apache.org/jira/browse/SPARK-18334
>             Project: Spark
>          Issue Type: Bug
>            Reporter: Yun Ni
>            Priority: Trivial
>
> MinHash currently is using the same `hashDistance` function as RandomProjection. This
does not make sense for MinHash because the Jaccard distance of two sets is not relevant to
the absolute distance of their hash buckets indices.
> This bug could affect accuracy of multi probing NN search for MinHash.



--
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


Mime
View raw message