C04-1156 |
algorithm KNOWA is an English/Italian
|
word aligner
|
, which relies mostly on information
|
D12-1078 |
and parsing might be fixed if
|
word aligner
|
and parser are not mutually independent
|
D09-1039 |
may be possible to integrate a
|
word aligner
|
or fragment aligner directly
|
C04-1156 |
present KNOWA , an English/Italian
|
word aligner
|
, developed at ITC-irst , which
|
D12-1078 |
the current state of the art ,
|
word aligner
|
and monolingual parser are trained
|
C04-1053 |
function words the performance of the
|
word aligner
|
improves in both precision and
|
D12-1078 |
although we do not invent a new
|
word aligner
|
exploiting syntactic information
|
C04-1053 |
with respect to statistics-based
|
word aligners
|
, which are expected to be able
|
D08-1092 |
features , which take counts from the
|
word aligner
|
's hard top-1 alignment output
|
C04-1053 |
original English texts . Then , the
|
word aligner
|
may align words incorrectly ,
|
D11-1082 |
power , but are easy to decode .
|
Word aligners
|
that model the alignment matrix
|
D08-1092 |
between v and v ' by an independent
|
word aligner
|
.2 Before defining alignment
|
C04-1156 |
algorithm . Also , knowledge-intensive
|
word aligners
|
may be more effective when word
|
D12-1078 |
. On the one hand , an average
|
word aligner
|
does not consider the syntax
|
C04-1053 |
availability of an English/Italian
|
word aligner
|
with very good performance in
|
C04-1053 |
same dictionaries used by the
|
word aligner
|
, and to maximize , whenever
|
D08-1092 |
probabilities were obtained from the
|
word aligner
|
of Liang et al. ( 2006 ) and
|
C04-1192 |
task evaluation ) of different
|
word aligners
|
( www.cs.unt.edu/~rada/wpt ,
|
C04-1156 |
compared the performances of two
|
word aligners
|
, one exclusively based on statistical
|
D09-1086 |
an automatic English parser and
|
word aligner
|
. The sourcelanguage English
|