as their new home is under contrib/queries.
This means contrib/similarity is not completely empty. I am leaving it in SVN
for now, as it seems like it could be a nice warm home for custom Similarity
implementations.
git-svn-id: https://svn.apache.org/repos/asf/lucene/java/trunk@480200 13f79535-47bb-0310-9956-ffa450edef68
was no point in explicitly looking for them as the scoring algorithm would effictively ignore them.
I did a test and indexed 700 pages on a corporate web site and then ran the MoreLikeThis code on them
and 1/2 of the docs had stop words identified as interesting.
So - I added code in to ignore stop words, but make it backward compatible so that by default this code
is not used.
git-svn-id: https://svn.apache.org/repos/asf/lucene/java/trunk@169512 13f79535-47bb-0310-9956-ffa450edef68
[2] Added convenience routine retrieveInterestingTerms() which makes it easier to get at the "interesting words" in a document.
git-svn-id: https://svn.apache.org/repos/asf/lucene/java/trunk@169508 13f79535-47bb-0310-9956-ffa450edef68
Otis Gospodnetic
Daniel Naber
Yonik Seeley
David Spencer
Erik Hatcher