[
https://issues.apache.org/jira/browse/LUCENE-1908?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12754615#action_12754615
]
Mark Miller commented on LUCENE-1908:
-------------------------------------
bq. In that work we got best results from Lucene
Thats funny - I didn't notice you were an author one that one! Small world.
The original idea of why the cosine normalization for the doc vector is bad, I
got from the free intro to info retrieval book thats out there -
but what it says doesn't fully jive with the info I am finding elsewhere, or my
own common sense. Thats what has me most confused
at the moment - the intro to ir book appears to break it down so that you can
explain it with the math (why going into the unit vector space favors longer
docs) - but other work I am seeing says
the math tells you no such thing, and its just comparing it to the computed
relevancy curve that tells you its not great.
I dunno :) - though at least I know a lot more than I did a few days ago - it
never even occurred to me how the scoring we did equated to any kind
of dot product before this - I used to read Lucene's scoring algorithm and then
look at the code and it was like .... okay - sure ... - so I've come a long way.
> Similarity javadocs for scoring function to relate more tightly to scoring
> models in effect
> -------------------------------------------------------------------------------------------
>
> Key: LUCENE-1908
> URL: https://issues.apache.org/jira/browse/LUCENE-1908
> Project: Lucene - Java
> Issue Type: Improvement
> Components: Search
> Reporter: Doron Cohen
> Assignee: Doron Cohen
> Priority: Minor
> Fix For: 2.9
>
> Attachments: LUCENE-1908.patch, LUCENE-1908.patch
>
>
> See discussion in the related issue.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
