#9601We present a framework for word alignment based on log-linear models.
tech,9-1-P05-1057,ak
<term>
word alignment
</term>
based on
<term>
log-linear models
</term>
. All
<term>
knowledge sources
</term>
#9605We present a framework for word alignment based on log-linear models.
other,1-2-P05-1057,ak
<term>
log-linear models
</term>
. All
<term>
knowledge sources
</term>
are treated as
<term>
feature functions
#9609All knowledge sources are treated as feature functions, which depend on the source language sentence, the target language sentence and possible additional variables.
tech,6-2-P05-1057,ak
knowledge sources
</term>
are treated as
<term>
feature functions
</term>
, which depend on the
<term>
source
#9614All knowledge sources are treated as feature functions, which depend on the source language sentence, the target language sentence and possible additional variables.
other,13-2-P05-1057,ak
functions
</term>
, which depend on the
<term>
source language sentence
</term>
, the
<term>
target language sentence
#9621All knowledge sources are treated as feature functions, which depend on the source language sentence, the target language sentence and possible additional variables.
other,18-2-P05-1057,ak
source language sentence
</term>
, the
<term>
target language sentence
</term>
and possible additional
<term>
variables
#9626All knowledge sources are treated as feature functions, which depend on the source language sentence, the target language sentence and possible additional variables.
other,24-2-P05-1057,ak
sentence
</term>
and possible additional
<term>
variables
</term>
.
<term>
Log-linear models
</term>
allow
#9632All knowledge sources are treated as feature functions, which depend on the source language sentence, the target language sentence and possible additional variables .
tech,0-3-P05-1057,ak
additional
<term>
variables
</term>
.
<term>
Log-linear models
</term>
allow
<term>
statistical alignment
#9634All knowledge sources are treated as feature functions, which depend on the source language sentence, the target language sentence and possible additional variables. Log-linear models allow statistical alignment models to be easily extended by incorporating syntactic information.
tech,3-3-P05-1057,ak
<term>
Log-linear models
</term>
allow
<term>
statistical alignment models
</term>
to be easily extended by incorporating
#9637Log-linear models allow statistical alignment models to be easily extended by incorporating syntactic information.
other,12-3-P05-1057,ak
easily extended by incorporating
<term>
syntactic information
</term>
. In this paper , we use
<term>
IBM
#9646Log-linear models allow statistical alignment models to be easily extended by incorporating syntactic information.
model,6-4-P05-1057,ak
information
</term>
. In this paper , we use
<term>
IBM Model 3 alignment properties
</term>
,
<term>
POS correspondence
</term>
,
#9655In this paper, we use IBM Model 3 alignment properties, POS correspondence, and bilingual dictionary coverage as features.
model,12-4-P05-1057,ak
Model 3 alignment properties
</term>
,
<term>
POS correspondence
</term>
, and
<term>
bilingual dictionary coverage
#9661In this paper, we use IBM Model 3 alignment properties, POS correspondence, and bilingual dictionary coverage as features.
model,16-4-P05-1057,ak
<term>
POS correspondence
</term>
, and
<term>
bilingual dictionary coverage
</term>
as
<term>
features
</term>
. Our experiments
#9665In this paper, we use IBM Model 3 alignment properties, POS correspondence, and bilingual dictionary coverage as features.
other,20-4-P05-1057,ak
bilingual dictionary coverage
</term>
as
<term>
features
</term>
. Our experiments show that
<term>
#9669In this paper, we use IBM Model 3 alignment properties, POS correspondence, and bilingual dictionary coverage as features .
tech,4-5-P05-1057,ak
</term>
. Our experiments show that
<term>
log-linear models
</term>
significantly outperform
<term>
IBM
#9675Our experiments show that log-linear models significantly outperform IBM translation models.
tech,8-5-P05-1057,ak
models
</term>
significantly outperform
<term>
IBM translation models
</term>
. This paper proposes an
<term>
alignment
#9679Our experiments show that log-linear models significantly outperform IBM translation models.