W10-4161 |
focus on how to apply CRFs in our
|
person name recognition
|
task . 3.1 CRFs-based name recognition
|
W10-4154 |
whole task into two subtasks : 1 .
|
Person name recognition
|
. It includes : 1.1 Distinguish
|
W02-1817 |
surnames or titles often trigger
|
person name recognition
|
on the following 2 or more characters
|
W06-0123 |
, to keep the high accuracy of
|
person name recognition
|
, the features about organization
|
W10-4161 |
used in each document . Then ,
|
person name recognition
|
is processing . The training
|
W03-1509 |
in the surname list can trigger
|
person name recognition
|
. A list of person title list
|
W06-0130 |
post-processing improved the performance of
|
person name recognition
|
, but it degraded the performance
|
W02-1817 |
avoid apparent mistakes . As to
|
person name recognition
|
, there are several strict restrictions
|
W07-0803 |
developed a tool for multilingual
|
person name recognition
|
that focuses on the " Who " part
|
W10-4161 |
names caused by sparse data . 3
|
Person Name Recognition
|
In this section , we focus on
|
W10-4161 |
present , the main technology of
|
person name recognition
|
is used statistical models ,
|
W03-1509 |
is the human knowledge used for
|
person name recognition
|
and the method of how to incorporate
|
I05-3028 |
context . We defined 17 roles for
|
person name recognition
|
and 10 roles for location name
|
W02-1817 |
The precision rate of Chinese
|
person names recognition
|
can achieve over 95 % after POS
|
W10-4161 |
corpus . 5.1 Preprocessing and
|
Person Name Recognition
|
Firstly , a word segmentation
|
W06-0110 |
statistical models for Chinese
|
person names recognition
|
and Chinese location names recognition
|
W06-0123 |
features to improve the accuracy of
|
person name recognition
|
, which are CNn ( n = -2 , -1,0,1,2
|
P98-1036 |
BBN ( Weischedel , 1995 ) on the
|
person name recognition
|
portion have very high recall
|
W03-1718 |
of 68.02 % for single character
|
person name recognition
|
. An alternative view of the
|
C04-1066 |
91.65 % recall for the Chinese
|
person names recognition
|
in the People 's Daily . Goh
|