Grant Ingersoll wrote:
I believe there is a MoreLikeThis class floating around somewhere (I
think it is in the contrib/similarity package). The Lucene book also
has a good example, and I have some examples at
http://www.cnlp.org/apachecon2005 that demonstrate using term vectors
to do this
Klaus wrote:
Hi,
is there are build-in method for finding similar documents to one given
document?
Thx,
Klaus
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
I've implemented a simple relevance feedback algorithm which extracts
terms from all interesting documents and builds up a new query with this
terms. This is pretty simple but It works in most cases.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
