other,32-3-A94-1011,bq |
notions of
<term>
noun group
</term>
,
<term>
|
verb group
|
</term>
, and so on which is inherently extensible
|
#19977
A novel method for adding linguistic annotation to corpora is presented which involves using a statistical POS tagger in conjunction with unsupervised structure finding methods to derive notions of noun group,verb group, and so on which is inherently extensible to more sophisticated annotation, and does not require a pre-tagged corpus to fit. |
tech,18-9-A94-1011,bq |
check
</term>
, which is creating a
<term>
|
PC based tool
|
</term>
to be used in the
<term>
technical
|
#20181
This paper reports on work done for the LRE project SmTA double check, which is creating aPC based tool to be used in the technical abstracting industry. |
measure(ment),12-6-A94-1011,bq |
high-precision
<term>
rules
</term>
can result in a
<term>
|
low precision
|
</term>
when used together , and develop
|
#20081
We investigate how sets of individually high-precision rules can result in alow precision when used together, and develop some theory about these probably-correct rules. |
other,20-5-A94-1011,bq |
from the
<term>
weighted sum
</term>
of a
<term>
|
word set representation
|
</term>
. We investigate how sets of individually
|
#20065
This leads us to consider the assignment of descriptors from individual phrases rather than from the weighted sum of aword set representation. |
other,15-4-A94-1011,bq |
representation of documents
</term>
over a
<term>
|
word set based representation
|
</term>
of them is that
<term>
linguistically
|
#20017
One of the distinguishing features of a more linguistically sophisticated representation of documents over aword set based representation of them is that linguistically sophisticated units are more frequently individually good predictors of document descriptors (keywords) than single words are. |
lr,52-3-A94-1011,bq |
annotation
</term>
, and does not require a
<term>
|
pre-tagged corpus
|
</term>
to fit . One of the distinguishing
|
#19997
A novel method for adding linguistic annotation to corpora is presented which involves using a statistical POS tagger in conjunction with unsupervised structure finding methods to derive notions of noun group, verb group, and so on which is inherently extensible to more sophisticated annotation, and does not require apre-tagged corpus to fit. |
tech,26-7-A94-1011,bq |
representations
</term>
, and show that a
<term>
|
statistically fitted rule-based model
|
</term>
provides significantly improved performance
|
#20122
We then proceed to repeat results which show that standard statistical models are not particularly suitable for exploiting linguistically sophisticated representations, and show that astatistically fitted rule-based model provides significantly improved performance for sophisticated representations. |
tech,15-3-A94-1011,bq |
is presented which involves using a
<term>
|
statistical POS tagger
|
</term>
in conjunction with
<term>
unsupervised
|
#19960
A novel method for adding linguistic annotation to corpora is presented which involves using astatistical POS tagger in conjunction with unsupervised structure finding methods to derive notions of noun group, verb group, and so on which is inherently extensible to more sophisticated annotation, and does not require a pre-tagged corpus to fit. |
other,5-3-A94-1011,bq |
</term>
. A novel method for adding
<term>
|
linguistic annotation
|
</term>
to
<term>
corpora
</term>
is presented
|
#19950
A novel method for addinglinguistic annotation to corpora is presented which involves using a statistical POS tagger in conjunction with unsupervised structure finding methods to derive notions of noun group, verb group, and so on which is inherently extensible to more sophisticated annotation, and does not require a pre-tagged corpus to fit. |
tech,16-2-A94-1011,bq |
if the power of recently developed
<term>
|
NLP techniques
|
</term>
are to be successfully applied in
|
#19935
This perplexing fact needs both an explanation and a solution if the power of recently developedNLP techniques are to be successfully applied in IR. |
other,8-8-A94-1011,bq |
statistical systems
</term>
can exploit
<term>
|
sophisticated representations of documents
|
</term>
, and lends some support to the use
|
#20142
It therefore shows that statistical systems can exploitsophisticated representations of documents, and lends some support to the use of more linguistically sophisticated representations for document classification. |
other,18-7-A94-1011,bq |
particularly suitable for exploiting
<term>
|
linguistically sophisticated representations
|
</term>
, and show that a
<term>
statistically
|
#20114
We then proceed to repeat results which show that standard statistical models are not particularly suitable for exploitinglinguistically sophisticated representations, and show that a statistically fitted rule-based model provides significantly improved performance for sophisticated representations. |
tech,26-8-A94-1011,bq |
sophisticated representations
</term>
for
<term>
|
document classification
|
</term>
. This paper reports on work done
|
#20160
It therefore shows that statistical systems can exploit sophisticated representations of documents, and lends some support to the use of more linguistically sophisticated representations fordocument classification. |
tech,6-1-A94-1011,bq |
use of
<term>
NLP techniques
</term>
for
<term>
|
document classification
|
</term>
has not produced significant improvements
|
#19889
The use of NLP techniques fordocument classification has not produced significant improvements in performance within the standard term weighting statistical assignment paradigm (Fagan 1987; Lewis, 1992bc; Buckley, 1993). |
other,7-6-A94-1011,bq |
sets of individually high-precision
<term>
|
rules
|
</term>
can result in a
<term>
low precision
|
#20076
We investigate how sets of individually high-precisionrules can result in a low precision when used together, and develop some theory about these probably-correct rules. |
tech,24-2-A94-1011,bq |
</term>
are to be successfully applied in
<term>
|
IR
|
</term>
. A novel method for adding
<term>
|
#19943
This perplexing fact needs both an explanation and a solution if the power of recently developed NLP techniques are to be successfully applied inIR. |
other,11-5-A94-1011,bq |
<term>
descriptors
</term>
from individual
<term>
|
phrases
|
</term>
rather than from the
<term>
weighted
|
#20056
This leads us to consider the assignment of descriptors from individualphrases rather than from the weighted sum of a word set representation. |
other,8-4-A94-1011,bq |
distinguishing features of a more
<term>
|
linguistically sophisticated representation of documents
|
</term>
over a
<term>
word set based representation
|
#20010
One of the distinguishing features of a morelinguistically sophisticated representation of documents over a word set based representation of them is that linguistically sophisticated units are more frequently individually good predictors of document descriptors (keywords) than single words are. |
other,22-8-A94-1011,bq |
lends some support to the use of more
<term>
|
linguistically sophisticated representations
|
</term>
for
<term>
document classification
</term>
|
#20156
It therefore shows that statistical systems can exploit sophisticated representations of documents, and lends some support to the use of morelinguistically sophisticated representations for document classification. |
other,8-5-A94-1011,bq |
leads us to consider the assignment of
<term>
|
descriptors
|
</term>
from individual
<term>
phrases
</term>
|
#20053
This leads us to consider the assignment ofdescriptors from individual phrases rather than from the weighted sum of a word set representation. |