tech,10-4-N04-4028,bq |
The
<term>
information extraction system
</term>
we evaluate is based on a
<term>
linear-chain conditional random field ( CRF )
</term>
, a
<term>
probabilistic model
</term>
which has performed well on
<term>
information extraction tasks
</term>
because of its ability to capture arbitrary , overlapping
<term>
features
</term>
of the input in a
<term>
Markov model
</term>
.
|
#6821
The information extraction system we evaluate is based on alinear-chain conditional random field ( CRF ), a probabilistic model which has performed well on information extraction tasks because of its ability to capture arbitrary, overlapping features of the input in a Markov model. |
other,14-5-N04-4028,bq |
We implement several techniques to estimate the
<term>
confidence
</term>
of both
<term>
extracted fields
</term>
and entire
<term>
multi-field records
</term>
, obtaining an
<term>
average precision
</term>
of 98 % for retrieving correct
<term>
fields
</term>
and 87 % for
<term>
multi-field records
</term>
.
|
#6872
We implement several techniques to estimate the confidence of both extracted fields and entiremulti-field records, obtaining an average precision of 98% for retrieving correct fields and 87% for multi-field records. |
other,8-1-N04-4028,bq |
<term>
Information extraction techniques
</term>
automatically create
<term>
structured databases
</term>
from
<term>
unstructured data sources
</term>
, such as the
<term>
Web
</term>
or
<term>
newswire documents
</term>
.
|
#6762
Information extraction techniques automatically create structured databases fromunstructured data sources, such as the Web or newswire documents. |
other,17-1-N04-4028,bq |
<term>
Information extraction techniques
</term>
automatically create
<term>
structured databases
</term>
from
<term>
unstructured data sources
</term>
, such as the
<term>
Web
</term>
or
<term>
newswire documents
</term>
.
|
#6771
Information extraction techniques automatically create structured databases from unstructured data sources, such as the Web ornewswire documents. |
other,21-3-N04-4028,bq |
For many reasons , it is highly desirable to accurately estimate the
<term>
confidence
</term>
the system has in the correctness of each
<term>
extracted field
</term>
.
|
#6808
For many reasons, it is highly desirable to accurately estimate the confidence the system has in the correctness of eachextracted field. |
model,44-4-N04-4028,bq |
The
<term>
information extraction system
</term>
we evaluate is based on a
<term>
linear-chain conditional random field ( CRF )
</term>
, a
<term>
probabilistic model
</term>
which has performed well on
<term>
information extraction tasks
</term>
because of its ability to capture arbitrary , overlapping
<term>
features
</term>
of the input in a
<term>
Markov model
</term>
.
|
#6855
The information extraction system we evaluate is based on a linear-chain conditional random field (CRF), a probabilistic model which has performed well on information extraction tasks because of its ability to capture arbitrary, overlapping features of the input in aMarkov model. |
other,32-5-N04-4028,bq |
We implement several techniques to estimate the
<term>
confidence
</term>
of both
<term>
extracted fields
</term>
and entire
<term>
multi-field records
</term>
, obtaining an
<term>
average precision
</term>
of 98 % for retrieving correct
<term>
fields
</term>
and 87 % for
<term>
multi-field records
</term>
.
|
#6890
We implement several techniques to estimate the confidence of both extracted fields and entire multi-field records, obtaining an average precision of 98% for retrieving correct fields and 87% formulti-field records. |
other,27-5-N04-4028,bq |
We implement several techniques to estimate the
<term>
confidence
</term>
of both
<term>
extracted fields
</term>
and entire
<term>
multi-field records
</term>
, obtaining an
<term>
average precision
</term>
of 98 % for retrieving correct
<term>
fields
</term>
and 87 % for
<term>
multi-field records
</term>
.
|
#6885
We implement several techniques to estimate the confidence of both extracted fields and entire multi-field records, obtaining an average precision of 98% for retrieving correctfields and 87% for multi-field records. |
other,10-5-N04-4028,bq |
We implement several techniques to estimate the
<term>
confidence
</term>
of both
<term>
extracted fields
</term>
and entire
<term>
multi-field records
</term>
, obtaining an
<term>
average precision
</term>
of 98 % for retrieving correct
<term>
fields
</term>
and 87 % for
<term>
multi-field records
</term>
.
|
#6868
We implement several techniques to estimate the confidence of bothextracted fields and entire multi-field records, obtaining an average precision of 98% for retrieving correct fields and 87% for multi-field records. |
tech,0-1-N04-4028,bq |
Results indicate that the system yields higher performance than a
<term>
baseline
</term>
on all three aspects .
<term>
Information extraction techniques
</term>
automatically create
<term>
structured databases
</term>
from
<term>
unstructured data sources
</term>
, such as the
<term>
Web
</term>
or
<term>
newswire documents
</term>
.
|
#6754
Results indicate that the system yields higher performance than a baseline on all three aspects.Information extraction techniques automatically create structured databases from unstructured data sources, such as the Web or newswire documents. |
other,26-4-N04-4028,bq |
The
<term>
information extraction system
</term>
we evaluate is based on a
<term>
linear-chain conditional random field ( CRF )
</term>
, a
<term>
probabilistic model
</term>
which has performed well on
<term>
information extraction tasks
</term>
because of its ability to capture arbitrary , overlapping
<term>
features
</term>
of the input in a
<term>
Markov model
</term>
.
|
#6837
The information extraction system we evaluate is based on a linear-chain conditional random field (CRF), a probabilistic model which has performed well oninformation extraction tasks because of its ability to capture arbitrary, overlapping features of the input in a Markov model. |
other,5-1-N04-4028,bq |
<term>
Information extraction techniques
</term>
automatically create
<term>
structured databases
</term>
from
<term>
unstructured data sources
</term>
, such as the
<term>
Web
</term>
or
<term>
newswire documents
</term>
.
|
#6759
Information extraction techniques automatically createstructured databases from unstructured data sources, such as the Web or newswire documents. |
measure(ment),19-5-N04-4028,bq |
We implement several techniques to estimate the
<term>
confidence
</term>
of both
<term>
extracted fields
</term>
and entire
<term>
multi-field records
</term>
, obtaining an
<term>
average precision
</term>
of 98 % for retrieving correct
<term>
fields
</term>
and 87 % for
<term>
multi-field records
</term>
.
|
#6877
We implement several techniques to estimate the confidence of both extracted fields and entire multi-field records, obtaining anaverage precision of 98% for retrieving correct fields and 87% for multi-field records. |
other,7-5-N04-4028,bq |
We implement several techniques to estimate the
<term>
confidence
</term>
of both
<term>
extracted fields
</term>
and entire
<term>
multi-field records
</term>
, obtaining an
<term>
average precision
</term>
of 98 % for retrieving correct
<term>
fields
</term>
and 87 % for
<term>
multi-field records
</term>
.
|
#6865
We implement several techniques to estimate theconfidence of both extracted fields and entire multi-field records, obtaining an average precision of 98% for retrieving correct fields and 87% for multi-field records. |
other,38-4-N04-4028,bq |
The
<term>
information extraction system
</term>
we evaluate is based on a
<term>
linear-chain conditional random field ( CRF )
</term>
, a
<term>
probabilistic model
</term>
which has performed well on
<term>
information extraction tasks
</term>
because of its ability to capture arbitrary , overlapping
<term>
features
</term>
of the input in a
<term>
Markov model
</term>
.
|
#6849
The information extraction system we evaluate is based on a linear-chain conditional random field (CRF), a probabilistic model which has performed well on information extraction tasks because of its ability to capture arbitrary, overlappingfeatures of the input in a Markov model. |
other,15-1-N04-4028,bq |
<term>
Information extraction techniques
</term>
automatically create
<term>
structured databases
</term>
from
<term>
unstructured data sources
</term>
, such as the
<term>
Web
</term>
or
<term>
newswire documents
</term>
.
|
#6769
Information extraction techniques automatically create structured databases from unstructured data sources, such as theWeb or newswire documents. |
other,12-3-N04-4028,bq |
For many reasons , it is highly desirable to accurately estimate the
<term>
confidence
</term>
the system has in the correctness of each
<term>
extracted field
</term>
.
|
#6799
For many reasons, it is highly desirable to accurately estimate theconfidence the system has in the correctness of each extracted field. |
tech,1-4-N04-4028,bq |
The
<term>
information extraction system
</term>
we evaluate is based on a
<term>
linear-chain conditional random field ( CRF )
</term>
, a
<term>
probabilistic model
</term>
which has performed well on
<term>
information extraction tasks
</term>
because of its ability to capture arbitrary , overlapping
<term>
features
</term>
of the input in a
<term>
Markov model
</term>
.
|
#6812
Theinformation extraction system we evaluate is based on a linear-chain conditional random field (CRF), a probabilistic model which has performed well on information extraction tasks because of its ability to capture arbitrary, overlapping features of the input in a Markov model. |
tech,19-4-N04-4028,bq |
The
<term>
information extraction system
</term>
we evaluate is based on a
<term>
linear-chain conditional random field ( CRF )
</term>
, a
<term>
probabilistic model
</term>
which has performed well on
<term>
information extraction tasks
</term>
because of its ability to capture arbitrary , overlapping
<term>
features
</term>
of the input in a
<term>
Markov model
</term>
.
|
#6830
The information extraction system we evaluate is based on a linear-chain conditional random field (CRF), aprobabilistic model which has performed well on information extraction tasks because of its ability to capture arbitrary, overlapping features of the input in a Markov model. |
measure(ment),7-2-N04-4028,bq |
Despite the successes of these systems ,
<term>
accuracy
</term>
will always be imperfect .
|
#6781
Despite the successes of these systems,accuracy will always be imperfect. |