P06-1061 |
significantly improved by our segment
|
HMM IE system
|
. The important improvement from
|
P06-1061 |
extraction redundancy in the document
|
HMM IE system
|
from Section 2 . We calculate
|
P06-1061 |
confidence intervals ) of our Document
|
HMM IE system
|
( Doc HMM ) . The performance
|
P06-1061 |
criterion for evaluating our document
|
HMM IE systems
|
in Section 2 . Although it might
|
P06-1061 |
the SGT smoothing method to our
|
HMM IE systems
|
to alleviate the data sparseness
|
P06-1061 |
measure the performance of an
|
HMM IE system
|
. Document HMM IE modelling does
|
P06-1061 |
evaluating these document-based
|
HMM IE systems
|
. We used the same criterion
|
P06-1061 |
extraction performance of the segment
|
HMM IE system
|
. In the future , we intend to
|
P06-1061 |
important improvement from the segment
|
HMM IE system
|
that it has achieved zero extraction
|
P06-1061 |
system ( Doc HMM ) and the segment
|
HMM IE system
|
( Seg HMM ) . Compared to the
|
P06-1061 |
McCallum , 1999 ) for our document
|
HMM IE system
|
. An example of such an HMM is
|
P06-1061 |
performance than traditional document
|
HMM IE systems
|
. 2 Document-based HMM IE with
|
P06-1061 |
we train an HMM in the document
|
HMM IE system
|
from a set of training documents
|
P06-1061 |
possibly extracted by a document
|
HMM IE system
|
, for filling in the slot template
|
P06-1061 |
work We evaluated our document
|
HMM IE system
|
on the seminar announcements
|
P06-1061 |
comparison between the document
|
HMM IE system
|
( Doc HMM ) and the segment HMM
|
P06-1061 |
segments . The resulting segment
|
HMM IE system
|
using the segment retrieval method
|
P06-1061 |
better on all slots than their
|
HMM IE system
|
using absolute discounting .
|
P06-1061 |
an issue in the training for an
|
HMM IE system
|
, especially when more advanced
|
P06-1061 |
filler-containing text segments for a segment
|
HMM IE system
|
. <title> A DOM Tree Alignment
|