[ https://issues.apache.org/jira/browse/SOLR-13500?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16907000#comment-16907000 ]
Dr Oleg Savrasov commented on SOLR-13500: ----------------------------------------- At first glance this looks like duplicate of https://issues.apache.org/jira/browse/SOLR-12890 > Vector Search with Solr > ----------------------- > > Key: SOLR-13500 > URL: https://issues.apache.org/jira/browse/SOLR-13500 > Project: Solr > Issue Type: New Feature > Reporter: Dr Oleg Savrasov > Priority: Major > > In machine learning problems we usually need to find nearest neighbors in > dense high dimensional vector space. Often, this retrieval had to be > augmented with traditional Boolean retrieval. > While no efficient KNN implementation is available in Solr yet, in practice > people are using brute-force vector distance sorting after Boolean retrieval > limits number of candidate vectors. > It's proposed to add a clear way to implement such a solution with Solr. -- This message was sent by Atlassian JIRA (v7.6.14#76016) --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org