</term>
show that the best
<term>
accuracy
</term>
#28214Theparsing results show that the best accuracy is obtained using the grammar acquired from the same domain or the same class (fiction or nonfiction).
other,20-2-A97-1015,ak
paper , we report 1 ) Comparison of
<term>
structure distributions
</term>
across
<term>
domains
</term>
; 2 ) Examples
#28174In order to examine the domain dependence of parsing, in this paper, we report 1) Comparison ofstructure distributions across domains; 2) Examples of domain specific structures; and 3) Parsing experiment using some domain dependent grammars.
other,12-3-A97-1015,ak
</term>
and
<term>
idiosyncrasy
</term>
of
<term>
syntactic structure
</term>
. The
<term>
parsing results
</term>
#28210The observations using the Brown corpus demonstrate domain dependence and idiosyncrasy ofsyntactic structure.
other,31-1-A97-1015,ak
</term>
, in particular , the notion of
<term>
text ' domain '
</term>
. In order to examine the
<term>
domain
#28149A major concern in corpus based approaches is that the applicability of the acquired knowledge may be limited by some feature of the corpus, in particular, the notion oftext ' domain '.
lr,13-5-A97-1015,ak
parsing accuracy
</term>
and the size of
<term>
training corpus
</term>
.
<term>
GLOSSER
</term>
is designed
#28254We will also discuss the relationship between parsing accuracy and the size oftraining corpus.