"Does the tf-idf vector represents one doc or set of docs?"
IDF is calculated across all docs that contain the term.
TF is calculated for a single document containing the term.
Each term of each doc will have its own tf-idf.
-- Jack Krupansky
-----Original Message-----
From: Allen
Sent: Tuesday, May 29, 2012 12:11 PM
To: solr-user@lucene.apache.org
Subject: TF-IDF vector
Hi List,
I am curious about the meaning of tf-idf vector after reading this
http://wiki.apache.org/solr/TermVectorComponent.
The tf flag returns me the tf vector for just one doc. The df flag
returns me the df vector of all the docs in the index.
Does the tf-idf vector represents one doc or set of docs?
Too, can I specify a subset of docs which the df vector is calculated
on rather than the entire set of docs?
