W99-0623 |
constituents appears in the output of the
|
constituent voting
|
technique using k parsers . Call
|
C00-1051 |
positive results ibr each . Simple
|
constituent voting
|
, however , does not accept parametric
|
W99-0623 |
the number of votes required by
|
constituent voting
|
is greater than half of the parsers
|
C00-1051 |
Henderson and Brill ( 1999 ) examined
|
constituent voting
|
and naive Bayes classifi ( :
|
W99-0623 |
include it . We call this technique
|
constituent voting
|
. We include a constituent in
|
W99-0623 |
Brackets Under certain conditions the
|
constituent voting
|
and naïve Bayes constituent
|
W99-0623 |
non-parametric counterpart . The
|
constituent voting
|
and naïve Bayes techniques
|
W05-1102 |
developed a similar measurement for
|
constituent voting
|
based on the level of consensus
|
W05-1102 |
discrepancy may be a quirk of the
|
constituent voting
|
algorithm that constituent-based
|
W05-1102 |
parse integration method called
|
constituent voting
|
, where a hybrid parse is produced
|
W05-1102 |
over hybridisation methods like
|
constituent voting
|
that the gaps in trees described
|
W99-0623 |
parser switching in precision .4
|
Constituent voting
|
gives the highest accuracy for
|
W99-0623 |
the best individual parser , and
|
constituent voting
|
is significantly better than
|
W99-0623 |
recall of similarity switching and
|
constituent voting
|
are both significantly better
|
W99-0623 |
parser , but the preci sion of the
|
constituent voting
|
technique was the only result
|
W05-1102 |
t-test over paired scores . 5.2
|
Constituent voting
|
Henderson ( 1999 ) reports good
|
A00-2005 |
( SteSt ) using the unweighted
|
constituent voting
|
scheme of Henderson and Brill
|