lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alexy Khrabrov <>
Subject applying cosine similarity directly
Date Sat, 12 Sep 2009 06:25:14 GMT
Given that I have a field for which term vector was computed and stored, and
that field is the text of a document, I'd like to rank a subset of such
documents by similarity to a given held-out document, or query, directly
using the cosine measure.  How can that be done without going through
creating a usual query?

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message