P94-1033 |
count . The precision rate of
|
compound extraction
|
will then be greatly improved
|
P94-1033 |
Cluster Classification Model for
|
Compound Extraction
|
The first step to extract compounds
|
C00-2116 |
about 90 % accuracy of Thai open
|
compound extraction
|
. However , the algorithm emphasizes
|
P94-1033 |
therefore used as a feature for
|
compound extraction
|
. Using both the mutual information
|
W00-1219 |
efficient and robust for Chinese
|
compounds extraction
|
. And , mutual information mainly
|
P94-1033 |
two supplementary features for
|
compound extraction
|
. Parts of Speech Part of speech
|
P94-1033 |
the words as a third feature for
|
compound extraction
|
. An automatic compound retrieval
|
P94-1033 |
of speech as the features for
|
compound extraction
|
. The problem is modeled as a
|
C96-2208 |
for data preparation and open
|
compound extraction
|
. The cornpetitive selection
|
C96-1006 |
. ( 1 ) Refinement of nominal
|
compound extraction
|
procedure : The simplified procedure
|
W00-1219 |
strict . This is very helpful for
|
compound extraction
|
. 3.2 Compounds Extraction from
|
P94-1033 |
both the recall and precision for
|
compound extraction
|
is improved . The simulation
|
D11-1060 |
Bootstrapping has been applied to noun
|
compound extraction
|
as well . For example , Kim and
|
W11-3508 |
1994 ) applied decision tree for
|
compound extraction
|
, but their supervised learning
|
W00-1219 |
two simple systems for Chinese
|
compound extraction
|
-- CXtract . CXtract uses predominantly
|
C96-1006 |
extraction ; refining the nominal
|
compound extraction
|
procedure will considerably improve
|
W00-1219 |
statistical approach to Chinese
|
compounds extraction
|
from very large corpora using
|
W00-1219 |
47.43 % . In this method , the
|
compound extraction
|
problem is formulated as classification
|
W00-1219 |
they use the method for English
|
compounds extraction
|
while we extract Chinese compounds
|
W00-1219 |
corpus are of interest to us in
|
compound extraction
|
. We use a criterion , called
|