[
https://issues.apache.org/jira/browse/SOLR-12890?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16948930#comment-16948930
]
Doug Turnbull commented on SOLR-12890:
--------------------------------------
[~solrtrey] these are good overviews of vector approaches, one thing to clarify
is the difference between matching and controlling the recall of the matched
results and scoring accuracy. For example the ES implementation was simply a
vector scoring approach directly doing distance measures in a reranking step.
Hangry is really focused on recall in a vector space, with rougher scoring
accuracy.
I'm also not sure there's just one solution. Like relevance in general, it
might be good for Lucene/Solr to support a variety of approaches for the needed
use case
> Vector Search in Solr (Umbrella Issue)
> --------------------------------------
>
> Key: SOLR-12890
> URL: https://issues.apache.org/jira/browse/SOLR-12890
> Project: Solr
> Issue Type: New Feature
> Reporter: mosh
> Priority: Major
>
> We have recently come across a need to index documents containing vectors
> using solr, and have even worked on a small POC. We used an URP to calculate
> the LSH(we chose to use the superbit algorithm, but the code is designed in a
> way the algorithm picked can be easily chagned), and stored the vector in
> either sparse or dense forms, in a binary field.
> Perhaps an addition of an LSH URP in conjunction with a query parser that
> uses the same properties to calculate LSH(or maybe ktree, or some other
> algorithm all together) should be considered as a Solr feature?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
