|
failed , and it involves the two stages
|
of
|
1 ) parsing a set of
<term>
phrases
</term>
|
#19404
Robust parsing is applied only after a full analysis has failed, and it involves the two stages of 1) parsing a set of phrases and clauses, and 2) gluing them together to obtain a single semantic frame encoding the full meaning of the sentence. |
|
trained with a little
<term>
corpus
</term>
|
of
|
100,000
<term>
words
</term>
, the system guesses
|
#11237
After several experiments, and trained with a little corpusof 100,000 words, the system guesses correctly not placing commas with a precision of 96% and a recall of 98%. |
|
the
<term>
collection
</term>
and distribution
|
of
|
12,000
<term>
utterances
</term>
of
<term>
spontaneous
|
#18593
We summarize the motivation for this effort, the goals, the implementation of a multi-site data collection paradigm, and the accomplishments of MADCOW in monitoring the collection and distribution of 12,000 utterances of spontaneous speech from five sites for use in a multi-site common evaluation of speech, natural language and spoken language |
|
construct a
<term>
corpus
</term>
consisting
|
of
|
126,610
<term>
sentences
</term>
. This paper
|
#11466
In our experiment, the method could construct a corpus consisting of 126,610 sentences. |
|
retrieval
</term>
indicates an improvement
|
of
|
22-38 % in
<term>
average precision
</term>
|
#4578
Task-based evaluation using Arabic information retrieval indicates an improvement of 22-38% in average precision over unstemmed text, and 96% of the performance of the proprietary stemmer above. |
|
classifies 73.2 % in a
<term>
German corpus
</term>
|
of
|
2.284
<term>
SRHs
</term>
as either coherent
|
#2522
An evaluation of our system against the annotated data shows that, it successfully classifies 73.2% in a German corpusof 2.284 SRHs as either coherent or incoherent (given a baseline of 54.55%). |
|
and sentence error rates
</term>
by a factor
|
of
|
2.5 and 1.6 , respectively , on the
<term>
|
#18763
Together with the use of a larger training set, these modifications combined to reduce the speech recognition word and sentence error rates by a factor of 2.5 and 1.6, respectively, on the October '91 test set. |
tech,13-3-C92-4199,bq |
word formation
</term>
,
<term>
identification
|
of
|
2-character and 3-character Chinese names
|
#18329
The proposed mechanism includes title-driven name recognition, adaptive dynamic word formation, identification of 2-character and 3-character Chinese names without title. |
|
questions
</term>
. We found a potential increase
|
of
|
35 % in
<term>
MRR
</term>
with respect to
|
#10787
We found a potential increase of 35% in MRR with respect to the original question. |
|
WSJ
</term>
, an
<term>
error reduction
</term>
|
of
|
4.4 % on the best previous single automatically
|
#3001
Using these ideas together, the resulting tagger gives a 97.24% accuracy on the Penn Treebank WSJ, an error reductionof 4.4% on the best previous single automatically learned tagging result. |
|
precision
</term>
of 70 % and a
<term>
recall
</term>
|
of
|
49 % in the task of placing
<term>
commas
|
#11272
It also gets a precision of 70% and a recallof 49% in the task of placing commas. |
|
i860 chip
</term>
, which provides a factor
|
of
|
5 speed-up over a
<term>
SUN 4
</term>
for
<term>
|
#16936
The speech-search algorithm is implemented on a board with a single Intel i860 chip, which provides a factor of 5 speed-up over a SUN 4 for straight C code. |
|
incoherent ( given a
<term>
baseline
</term>
|
of
|
54.55 % ) . We propose a new
<term>
phrase-based
|
#2534
An evaluation of our system against the annotated data shows that, it successfully classifies 73.2% in a German corpus of 2.284 SRHs as either coherent or incoherent (given a baselineof 54.55%). |
|
% . It also gets a
<term>
precision
</term>
|
of
|
70 % and a
<term>
recall
</term>
of 49 % in
|
#11266
It also gets a precisionof 70% and a recall of 49% in the task of placing commas. |
|
the
<term>
baseline model ’s score
</term>
|
of
|
88.2 % . The article also introduces a
|
#8857
The new model achieved 89.75% F-measure, a 13% relative decrease in F-measure error over the baseline model’s scoreof 88.2%. |
|
commas
</term>
with a
<term>
precision
</term>
|
of
|
96 % and a
<term>
recall
</term>
of 98 % .
|
#11251
After several experiments, and trained with a little corpus of 100,000 words, the system guesses correctly not placing commas with a precisionof 96% and a recall of 98%. |
|
precision
</term>
of 96 % and a
<term>
recall
</term>
|
of
|
98 % . It also gets a
<term>
precision
</term>
|
#11257
After several experiments, and trained with a little corpus of 100,000 words, the system guesses correctly not placing commas with a precision of 96% and a recallof 98%. |
|
obtaining an
<term>
average precision
</term>
|
of
|
98 % for retrieving correct
<term>
fields
|
#6879
We implement several techniques to estimate the confidence of both extracted fields and entire multi-field records, obtaining an average precisionof 98% for retrieving correct fields and 87% for multi-field records. |
|
Translations
</term>
are produced by means
|
of
|
a
<term>
beam-search decoder
</term>
. Experimental
|
#7409
Translations are produced by means of a beam-search decoder. |
|
context-dependent phonetic modelling
</term>
, the use
|
of
|
a
<term>
bigram language model
</term>
in conjunction
|
#18717
These include context-dependent phonetic modelling, the use of a bigram language model in conjunction with a probabilistic LR parser, and refinements made to the lexicon. |