lecture18-learning-ranking-handouts-6-per

Lecture18-learning-ranking-handouts-6-per

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: 
 0.1728
 0.2432
 0.2750
 LM
 0.1773
 0.2516
 0.2656
 SVM
 Disk
4‐5
 Disk
3
 LM
 SVM
 Disk
3
 0.1646
 0.2355
 0.2675
   At
best
the
results
are
about
equal
to
LM
   Actually
a
ligle
bit
below
   Paper’s
adver)sement:
Easy
to
add
more
features
   This
is
illustrated
on
a
homepage
finding
task
on
 WT10G:
   [email protected],
baseline
SVM
58%
   SVM
with
URL‐depth,
and
in‐link
features:[email protected] Introduc)on to Informa)on Retrieval Sec.
15.4.2
 Introduc)on to Informa)on Retrieval “Learning
to
rank”
 “Learning
to
rank”
   Classifica)on
probably
isn’t
the
right
way
to
think
 about
score
learning:
   Assume
a
number
of
categories
C of
relevance
 exist
   Classifica)on
problems:
Map
to
a
unordered
set
of
classes
   Regression
problems:
Map
to
a
real
value
  ...
View Full Document

This document was uploaded on 02/26/2014.

Ask a homework question - tutors are online