tech,10-2-C92-2068,bq |
We focus on one speed-up element in the design of
<term>
unification algorithms
</term>
: avoidance of
<term>
copying
</term>
of
<term>
unmodified subgraphs
</term>
.
|
#17968
We focus on one speed-up element in the design ofunification algorithms: avoidance of copying of unmodified subgraphs. |
other,32-3-C92-2068,bq |
We propose a method of attaining such a design through a method of
<term>
structure-sharing
</term>
which avoids
<term>
log ( d ) overheads
</term>
often associated with
<term>
structure-sharing of graphs
</term>
without any use of costly
<term>
dependency pointers
</term>
.
|
#18010
We propose a method of attaining such a design through a method of structure-sharing which avoids log(d) overheads often associated with structure-sharing of graphs without any use of costlydependency pointers. |
tech,8-1-C92-2068,bq |
<term>
Graph unification
</term>
remains the most expensive part of
<term>
unification-based grammar parsing
</term>
.
|
#17954
Graph unification remains the most expensive part ofunification-based grammar parsing. |
other,17-2-C92-2068,bq |
We focus on one speed-up element in the design of
<term>
unification algorithms
</term>
: avoidance of
<term>
copying
</term>
of
<term>
unmodified subgraphs
</term>
.
|
#17975
We focus on one speed-up element in the design of unification algorithms: avoidance of copying ofunmodified subgraphs. |
other,15-4-C92-2068,bq |
The proposed scheme eliminates
<term>
redundant copying
</term>
while maintaining the
<term>
quasi-destructive scheme 's ability
</term>
to avoid
<term>
over copying
</term>
and
<term>
early copying
</term>
combined with its ability to handle
<term>
cyclic structures
</term>
without algorithmic additions .
|
#18028
The proposed scheme eliminates redundant copying while maintaining the quasi-destructive scheme's ability to avoidover copying and early copying combined with its ability to handle cyclic structures without algorithmic additions. |
other,18-4-C92-2068,bq |
The proposed scheme eliminates
<term>
redundant copying
</term>
while maintaining the
<term>
quasi-destructive scheme 's ability
</term>
to avoid
<term>
over copying
</term>
and
<term>
early copying
</term>
combined with its ability to handle
<term>
cyclic structures
</term>
without algorithmic additions .
|
#18031
The proposed scheme eliminates redundant copying while maintaining the quasi-destructive scheme's ability to avoid over copying andearly copying combined with its ability to handle cyclic structures without algorithmic additions. |
tech,15-2-C92-2068,bq |
We focus on one speed-up element in the design of
<term>
unification algorithms
</term>
: avoidance of
<term>
copying
</term>
of
<term>
unmodified subgraphs
</term>
.
|
#17973
We focus on one speed-up element in the design of unification algorithms: avoidance ofcopying of unmodified subgraphs. |
other,16-3-C92-2068,bq |
We propose a method of attaining such a design through a method of
<term>
structure-sharing
</term>
which avoids
<term>
log ( d ) overheads
</term>
often associated with
<term>
structure-sharing of graphs
</term>
without any use of costly
<term>
dependency pointers
</term>
.
|
#17994
We propose a method of attaining such a design through a method of structure-sharing which avoidslog ( d ) overheads often associated with structure-sharing of graphs without any use of costly dependency pointers. |
tech,0-1-C92-2068,bq |
The
<term>
accuracy rate
</term>
of
<term>
syntactic disambiguation
</term>
is raised from 46.0 % to 60.62 % by using this novel approach .
<term>
Graph unification
</term>
remains the most expensive part of
<term>
unification-based grammar parsing
</term>
.
|
#17946
The accuracy rate of syntactic disambiguation is raised from 46.0% to 60.62% by using this novel approach.Graph unification remains the most expensive part of unification-based grammar parsing. |
other,9-4-C92-2068,bq |
The proposed scheme eliminates
<term>
redundant copying
</term>
while maintaining the
<term>
quasi-destructive scheme 's ability
</term>
to avoid
<term>
over copying
</term>
and
<term>
early copying
</term>
combined with its ability to handle
<term>
cyclic structures
</term>
without algorithmic additions .
|
#18022
The proposed scheme eliminates redundant copying while maintaining thequasi-destructive scheme 's ability to avoid over copying and early copying combined with its ability to handle cyclic structures without algorithmic additions. |
tech,13-3-C92-2068,bq |
We propose a method of attaining such a design through a method of
<term>
structure-sharing
</term>
which avoids
<term>
log ( d ) overheads
</term>
often associated with
<term>
structure-sharing of graphs
</term>
without any use of costly
<term>
dependency pointers
</term>
.
|
#17991
We propose a method of attaining such a design through a method ofstructure-sharing which avoids log(d) overheads often associated with structure-sharing of graphs without any use of costly dependency pointers. |
other,4-4-C92-2068,bq |
The proposed scheme eliminates
<term>
redundant copying
</term>
while maintaining the
<term>
quasi-destructive scheme 's ability
</term>
to avoid
<term>
over copying
</term>
and
<term>
early copying
</term>
combined with its ability to handle
<term>
cyclic structures
</term>
without algorithmic additions .
|
#18017
The proposed scheme eliminatesredundant copying while maintaining the quasi-destructive scheme's ability to avoid over copying and early copying combined with its ability to handle cyclic structures without algorithmic additions. |
other,26-4-C92-2068,bq |
The proposed scheme eliminates
<term>
redundant copying
</term>
while maintaining the
<term>
quasi-destructive scheme 's ability
</term>
to avoid
<term>
over copying
</term>
and
<term>
early copying
</term>
combined with its ability to handle
<term>
cyclic structures
</term>
without algorithmic additions .
|
#18039
The proposed scheme eliminates redundant copying while maintaining the quasi-destructive scheme's ability to avoid over copying and early copying combined with its ability to handlecyclic structures without algorithmic additions. |
tech,24-3-C92-2068,bq |
We propose a method of attaining such a design through a method of
<term>
structure-sharing
</term>
which avoids
<term>
log ( d ) overheads
</term>
often associated with
<term>
structure-sharing of graphs
</term>
without any use of costly
<term>
dependency pointers
</term>
.
|
#18002
We propose a method of attaining such a design through a method of structure-sharing which avoids log(d) overheads often associated withstructure-sharing of graphs without any use of costly dependency pointers. |