|
</term>
and better
<term>
word similarity
</term>
|
performance
|
. The work presented in this paper is the
|
#5377
Finally, a novel feature weighting and selection function is presented, which yields superior feature vectors and better word similarityperformance. |
|
discussions
</term>
, and evaluated their
|
performance
|
by means of two experiments : coarse-level
|
#5466
We tested the clustering and filtering processes on electronic newsgroup discussions, and evaluated their performance by means of two experiments: coarse-level clustering and simple information retrieval. |
measure(ment),16-2-N04-1022,bq |
functions
</term>
that measure
<term>
translation
|
performance
|
</term>
. We describe a hierarchy of
<term>
|
#6573
This statistical approach aims to minimize expected loss of translation errors under loss functions that measure translation performance. |
|
language sentences
</term>
. We report the
|
performance
|
of the
<term>
MBR decoders
</term>
on a
<term>
|
#6615
We report the performance of the MBR decoders on a Chinese-to-English translation task. |
|
used to tune
<term>
statistical MT
</term>
|
performance
|
for specific
<term>
loss functions
</term>
|
#6639
Our results show that MBR decoding can be used to tune statistical MTperformance for specific loss functions. |
|
Results indicate that the system yields higher
|
performance
|
than a
<term>
baseline
</term>
on all three
|
#6745
Results indicate that the system yields higher performance than a baseline on all three aspects. |
measure(ment),26-2-H05-1012,bq |
unsupervised methods
</term>
yields superior
<term>
|
performance
|
</term>
. The
<term>
probabilistic model
</term>
|
#7292
We demonstrate that it is feasible to create training material for problems in machine translation and that a mixture of supervised and unsupervised methods yields superiorperformance. |
measure(ment),7-5-H05-1012,bq |
is contrasted with
<term>
human annotation
|
performance
|
</term>
. This paper presents a
<term>
phrase-based
|
#7335
Performance of the algorithm is contrasted with human annotation performance. |
measure(ment),23-1-I05-2021,bq |
directly on
<term>
word sense disambiguation
|
performance
|
</term>
, using standard
<term>
WSD evaluation
|
#7812
We present the first known empirical test of an increasingly common speculative claim, by evaluating a representative Chinese-to-English SMT model directly on word sense disambiguation performance, using standard WSD evaluation methodology and datasets from the Senseval-3 Chinese lexical sample task. |
|
data
</term>
by showing that it improves the
|
performance
|
of a state-of-the-art
<term>
statistical
|
#9059
We evaluate the quality of the extracted data by showing that it improves the performance of a state-of-the-art statistical machine translation system. |
measure(ment),6-4-E06-1035,bq |
</term>
. We then explore the impact on
<term>
|
performance
|
</term>
of using
<term>
ASR output
</term>
as
|
#10515
We then explore the impact onperformance of using ASR output as opposed to human transcription. |
|
discuss its application , and evaluate its
|
performance
|
. State-of-the-art
<term>
Question Answering
|
#10727
We describe a clustering algorithm which is sufficiently general to be applied to these diverse problems, discuss its application, and evaluate its performance. |
|
</term>
and evaluating
<term>
QA system
</term>
|
performance
|
on
<term>
paraphrased questions
</term>
. We
|
#10777
We investigate that claim by adopting a simple MT-based paraphrasing technique and evaluating QA systemperformance on paraphrased questions. |
other,8-1-P06-1013,bq |
an effective way of improving
<term>
system
|
performance
|
</term>
. This paper examines the benefits
|
#10977
Combination methods are an effective way of improving system performance. |
|
but claims that direct imitation of human
|
performance
|
is not the best way to implement many of
|
#12596
This paper defends that view, but claims that direct imitation of human performance is not the best way to implement many of these non-literal aspects of communication; that the new technology of powerful personal computers with integral graphics displays offers techniques superior to those of humans for these aspects, while still satisfying human communication needs. |
measure(ment),1-5-H90-1060,bq |
Resource Management corpus
</term>
. This
<term>
|
performance
|
</term>
is comparable to our best condition
|
#17102
Thisperformance is comparable to our best condition for this test suite, using 109 training speakers. |
|
the required
<term>
world knowledge
</term>
,
|
performance
|
degrades gracefully . Each of these techniques
|
#17523
For pragmatics processing, we describe how the method of abductive inference is inherently robust, in that an interpretation is always possible, so that in the absence of the required world knowledge, performance degrades gracefully. |
measure(ment),37-2-C92-1055,bq |
method
</term>
, fail to achieve high
<term>
|
performance
|
</term>
in real applications . The proposed
|
#17858
Owing to the problem of insufficient training data and approximation error introduced by the language model, traditional statistical approaches, which resolve ambiguities by indirectly and implicitly using maximum likelihood method, fail to achieve highperformance in real applications. |
|
<term>
constraint
</term>
for improving the
|
performance
|
of the
<term>
word-sense disambiguation algorithm
|
#19303
This result can be used as an additional source of constraint for improving the performance of the word-sense disambiguation algorithm. |
measure(ment),17-4-H92-1060,bq |
mechanism
</term>
through a breakdown of the
<term>
|
performance
|
</term>
of robustly parsed vs. fully parsed
|
#19452
We have assessed the degree of success of the robust parsing mechanism through a breakdown of theperformance of robustly parsed vs. fully parsed sentences on the October '91 dry-run test set. |