N07-1068 |
. For the language model , the
|
Jelinek-Mercer smoothing
|
method was employed with the
|
P11-1095 |
needs to tune one parameter : the
|
Jelinek-Mercer smoothing
|
parameter A used in the entity
|
W04-1104 |
be considered as an instance of
|
Jelinek-Mercer smoothing
|
. It is defined recursively as
|
P96-1041 |
This scheme is an instance of
|
Jelinek-Mercer smoothing
|
. Referring to equation ( 3 )
|
P11-1066 |
likelihood language model with
|
Jelinek-Mercer smoothing
|
can be D : ... for good cold
|
W09-2012 |
word to be predicted . A 5-gram
|
Jelinek-Mercer smoothing
|
language model on sentence x
|
W09-2012 |
Jelinek-Mercer " smoothing . As in
|
Jelinek-Mercer smoothing
|
( Jelinek and Mercer , 1980 )
|
W04-0307 |
smooth the probabilities using
|
Jelinek-Mercer smoothing
|
( Jelinek , 1997 ) , as described
|
W13-2504 |
techniques such as Backoff or
|
Jelinek-Mercer smoothing
|
, two techniques that generally
|
W13-2504 |
. We can also notice that the
|
Jelinek-Mercer smoothing
|
improves more notably the High-test
|
P96-1041 |
method , we use an instance of
|
Jelinek-Mercer smoothing
|
where we constrain all Ami-i
|
P96-1041 |
techniques , Katz smoothing and
|
Jelinek-Mercer smoothing
|
, perform consistently well across
|
P11-2005 |
1998 ) . We also compare against
|
Jelinek-Mercer smoothing
|
( JMLM ) , which interpolates
|
W13-2504 |
= ( r + 1 ) Nr +1 ( 6 ) Nr 2.3
|
Jelinek-Mercer Smoothing
|
As one alternative to missing
|
P96-1041 |
- niques , one a variation of
|
Jelinek-Mercer smoothing
|
and one a very simple linear
|
P14-2100 |
smoothing ( Ney et al. , 1995 ) ,
|
Jelinek-Mercer smoothing
|
( Jelinek and Mercer , 1980 )
|
P09-1082 |
et al. ( 2008 ) is that we use
|
Jelinek-Mercer smoothing
|
for equation 3 instead of Dirichlet
|
P96-1041 |
We implemented two versions of
|
Jelinek-Mercer smoothing
|
differing only in what data is
|
W13-2504 |
the Good-Turing estima - tions .
|
Jelinek-Mercer smoothing
|
counteracts the disadvantage
|
W07-0910 |
Lucene ( ILPS , 2005 ) and uses
|
Jelinek-Mercer smoothing
|
, controlled by the parameter
|