lecture18-learning-ranking-handouts-6-per

Idfscores disk45 wt10gweb 01785 02503 02666 01728

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: tures
   What
if
the
)tle
contains
some
but
not
all
query
terms
…
   Categorical
features
(query
terms
occur
in
plain,
boldface,
 italics,
etc)
   Scores
are
nonlinear
combina)ons
of
features
   Mul)level
relevance
judgments
(Perfect,
Good,
Fair,
 Bad,
etc)
   Complex
error
func)ons
   Not
always
a
unique,
easily
computable
sesng
of
 score
parameters
 Sec.
15.4.1
 Introduc)on to Informa)on Retrieval Introduc)on to Informa)on Retrieval Sec.
15.4.1
 A
richer
example
 Using
classifica)on
for
deciding
relevance
   Collect
a
training
corpus
of
(q, d, r)
triples
   Relevance
r is
s)ll
binary
for
now
   Document
is
represented
by
a
feature
vector
   A
linear
score
func)on
is
 Score(d, q) = Score(α, ω) = aα + bω + c   And
the
linear
classifier
is
 Decide
relevant
if
Score(d, q) > θ
   x
=
(α,
ω) 
α
is
cosine
similarity,
ω
is
minimum
query
window
size
   ω
is
the
the
shortest
text
span
that
includes
all
query
words
  ...
View Full Document

Ask a homework question - tutors are online