P06-1087 |
multiclass classification , we use
|
pairwise voting
|
. For all the reported experi
|
P03-1064 |
combined all their models with
|
pairwise voting
|
, yielding an accuracy of . The
|
W00-0730 |
class ( tag ) obtained through the
|
pairwise voting
|
. Since SVMs are vector based
|
P03-1064 |
generate the mapping table used in
|
pairwise voting
|
. The SNoW supertagger scanning
|
P13-2133 |
order of pref - erence . It is a
|
pairwise voting
|
, i.e. it compares every possible
|
J01-2002 |
with lexical items ) . When using
|
pairwise voting
|
on models trained using different
|
H05-1059 |
error reduction of supertagging by
|
pairwise voting
|
between left-to-right and right-toleft
|
P03-1064 |
Brill , 1995 ) . We use the same
|
pairwise voting
|
algorithm as in ( Chen et al.
|
W03-1728 |
Joshi , 2003 ) . In that paper ,
|
pairwise voting
|
( van Halteren et al. , 1998
|
P03-1064 |
of these two supertaggers with
|
pairwise voting
|
, we achieve an accuracy of ,
|
W03-1728 |
the opposite directions . The
|
pairwise voting
|
is not suitable in this application
|
P03-1064 |
Then we combine the results via
|
pairwise voting
|
as in ( van Halteren et al. ,
|
P98-1081 |
McNemar 's chi-square , p = 0 . 5
|
Pairwise Voting
|
So far , we have only used information
|
E99-1025 |
accuracy of a classifier , and
|
pairwise voting
|
. Pairwise voting works as follows
|
P98-1081 |
the best combi - nation . The
|
pairwise voting
|
system , using all four individual
|
E99-1025 |
classifier , and pairwise voting .
|
Pairwise voting
|
works as follows . First , for
|
W03-1728 |
not immediately follow an MM .
|
Pairwise voting
|
does not use any contextual information
|
P10-1068 |
e.g. , weighted majority vote ,
|
pairwise voting
|
( Halteren et al. , 1998 ) ,
|
N01-1025 |
the class obtained through the
|
pairwise voting
|
is used as the certain score
|
E99-1025 |
vote has an accuracy of 91.93 % .
|
Pairwise voting
|
yields an accuracy of 92.19 %
|