P06-1061 |
redundancy . Shown from Table 3 , the
|
segment retrieval
|
results have achieved high recall
|
P06-1061 |
extraction performance . Our current
|
segment retrieval
|
method requires the training
|
P06-1061 |
retrieval using HMMs We propose a
|
segment retrieval
|
approach for performing the first
|
P06-1061 |
have presented and evaluated our
|
segment retrieval
|
method . Document-wise retrieval
|
P06-1061 |
in Definition 4 to measure the
|
segment retrieval
|
accuracy . Definition 4 . Document-wise
|
P06-1061 |
process consists of two steps : a
|
segment retrieval
|
step followed by an extraction
|
P06-1061 |
evaluate average document-wise
|
segment retrieval
|
redundancy , as defined in Definition
|
P06-1061 |
learn an extraction-relevance
|
segment retrieval
|
system from the labelled texts
|
P06-1061 |
IE model , using the proposed
|
segment retrieval
|
approach , and comparing their
|
P06-1061 |
segment HMM IE system using the
|
segment retrieval
|
method has not only achieved
|
W05-0832 |
information that is useful in
|
segment retrieval
|
in the translation memory . For
|
P06-1061 |
performance Since the purpose of our
|
segment retrieval
|
is to identify relevant segments
|
P06-1061 |
observation sequences are used . 5.2
|
Segment retrieval
|
using HMMs After a retrieval
|
P06-1061 |
on the goodness of a particular
|
segment retrieval
|
method for our purpose : the
|
P13-1018 |
precision or for recall . 3 Parallel
|
Segment Retrieval
|
We will first abstract from the
|
P06-1061 |
imposing an extraction-relevant text
|
segment retrieval
|
in the segment HMM IE modelling
|
P06-1061 |
segments retrieved ) by a given
|
segment retrieval
|
system . We consider two criteria
|
P06-1061 |
extractor HMM . 5.1 Training HMMs for
|
segment retrieval
|
To train a retrieval HMM , it
|
P06-1061 |
most correctly filtered by the
|
segment retrieval
|
system only when all the extraction
|
P06-1061 |
sequence models used for IE .
|
Segment retrieval
|
for extraction is an important
|