lecture18-learning-ranking-handouts-6-per

Sigir2007 disconnuityexample

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Ordinal
regression
problems:
Map
to
an
ordered
set
of
 classes
   A
fairly
obscure
sub‐branch
of
sta)s)cs,
but
what
we
want
here
   This
formula)on
gives
extra
power:
   These
are
totally
ordered:
c1
<
c2
<
…
<
cJ
   This
is
the
ordinal
regression
setup
   Assume
training
data
is
available
consis)ng
of
 document‐query
pairs
represented
as
feature
 vectors
ψi
and
relevance
ranking
ci   Rela)ons
between
relevance
levels
are
modeled
   Documents
are
good
versus
other
documents
for
query
 given
collec)on;
not
an
absolute
scale
of
goodness
 4 5/30/11 Introduc)on to Informa)on Retrieval Sec.
15.4.1
 Introduc)on to Informa)on Retrieval Modified
example
 “Learning
to
rank”
   Collect
a
training
corpus
of
(q, d, r)
triples
   Relevance
r is
here
4
values
   Perfect,
Relevant,
Weak,
Nonrelevant
   We
could
do...
View Full Document

This document was uploaded on 02/26/2014.

Ask a homework question - tutors are online