[
https://issues.apache.org/jira/browse/LUCENE-8340?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16496509#comment-16496509
]
Adrien Grand commented on LUCENE-8340:
--------------------------------------
Here is a patch that implements such a query on LongPoint, since this is what
users would typically use to index dates. It also requires that the field has
doc values. It computes scores as {{pivotDistance / (distance +
pivotDistance)}} very similarly to {{FeatureField.newSaturationQuery}} and
computes the range that values must belong to in {{setMinCompetitiveScore}}
using points in order to efficiently skip non-competitive documents.
Currently it works on its own, but because it can't expose useful impacts (the
only thing it knows is that the maximum score is 1), we need to do some changes
to make boolean queries as well in order to propagate
{{setMinCompetitiveScore}} calls to their sub clauses so that we could also
skip non-competitive docs when boosting full-text matches by recency.
> Allow to boost by recency
> -------------------------
>
> Key: LUCENE-8340
> URL: https://issues.apache.org/jira/browse/LUCENE-8340
> Project: Lucene - Core
> Issue Type: Improvement
> Reporter: Adrien Grand
> Priority: Minor
> Attachments: LUCENE-8340.patch
>
>
> I would like that we support something like
> \{{FeatureField.newSaturationQuery}} but that works with features that are
> computed dynamically like recency or geo-distance, and is still optimized for
> top-hits collection. I'm starting with recency because it makes things a bit
> easier even though I suspect that geo-distance might be a more common need.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
