#506(iii) Rapid system development and porting to new domains via knowledge-based automated acquisition of grammars.
Listen-Communicate-Show ( LCS )
</term>
is a
new
paradigm for
<term>
human interaction with
#786Listen-Communicate-Show (LCS) is a new paradigm for human interaction with data sources.
developing applications of this technology in
new
domains . Recent advances in
<term>
Automatic
#908We have demonstrated this capability in several field exercises with the Marines and are currently developing applications of this technology in new domains.
recognition
</term>
has brought to light a
new
problem : as
<term>
dialog systems
</term>
#941However, the improved speech recognition has brought to light a new problem: as dialog systems understand more of what the user tells them, they need to be more sophisticated at responding to the user.
</term>
, a
<term>
sentence planner
</term>
, and a
new
methodology for automatically training
<term>
#1349In this paper, we present SPoT, a sentence planner, and a new methodology for automatically training SPoT on the basis of feedback provided by human judges.
baseline
</term>
of 54.55 % ) . We propose a
new
<term>
phrase-based translation model
</term>
#2543We propose a new phrase-based translation model and decoding algorithm that enables us to evaluate and compare several, previously proposed phrase-based translation models.
constraint-based parser/generator
</term>
. We present a
new
<term>
part-of-speech tagger
</term>
that demonstrates
#2913We present a new part-of-speech tagger that demonstrates the following ideas: (i) explicit use of both preceding and following tag contexts via a dependency network representation, (ii) broad use of lexical features, including jointly conditioning on multiple consecutive words, (iii) effective use of priors in conditional loglinear models, and (iv) fine-grained modeling of unknown word features.
predicate-argument structures
</term>
. We also introduce a
new
way of automatically identifying
<term>
predicate
#3734We also introduce a new way of automatically identifying predicate argument structures, which is central to our IE paradigm.
undisambiguated
<term>
corpus data
</term>
. We describe a
new
approach which involves clustering
<term>
#3906We describe a new approach which involves clustering subcategorization frame (SCF) distributions using the Information Bottleneck and nearest neighbour methods.
tech,17-1-P03-1030,ak
Detection and Tracking tasks
</term>
of
<term>
new
event detection
</term>
. In this paper we
#4060Link detection has been regarded as a core technology for the Topic Detection and Tracking tasks ofnew event detection.
tech,9-2-P03-1030,ak
<term>
story link detection
</term>
and
<term>
new
event detection
</term>
as
<term>
information
#4073In this paper we formulate story link detection andnew event detection as information retrieval task and hypothesize on the impact of precision and recall on both systems.
these arguments , we introduce a number of
new
performance enhancing techniques including
#4103Motivated by these arguments, we introduce a number of new performance enhancing techniques including part of speech tagging, new similarity measures and expanded stop lists.
including
<term>
part of speech tagging
</term>
,
new
<term>
similarity measures
</term>
and expanded
#4113Motivated by these arguments, we introduce a number of new performance enhancing techniques including part of speech tagging, new similarity measures and expanded stop lists.
algorithm
</term>
for automatically acquiring
new
<term>
stems
</term>
from a
<term>
155 million
#4722To improve the segmentation accuracy, we use an unsupervised algorithm for automatically acquiring new stems from a 155 million word unsegmented corpus, and re-estimate the model parameters with the expanded vocabulary and training corpus.
non-Bayesian models
</term>
. We describe a
new
method for the representation of NLP structures
#5433We describe a new method for the representation of NLP structures within reranking approaches.
building
<term>
translation systems
</term>
for
new
<term>
language pairs
</term>
or new
<term>
domains
#6800This is particularly important when building translation systems for new language pairs or new domains.
</term>
for new
<term>
language pairs
</term>
or
new
<term>
domains
</term>
. This workshop is intended
#6804This is particularly important when building translation systems for new language pairs or new domains.
features
</term>
into account . We introduce a
new
method for the
<term>
reranking task
</term>
#8127We introduce a new method for the reranking task, based on the boosting approach to ranking problems described in Freund et al. (1998).
included in the original
<term>
model
</term>
. The
new
<term>
model
</term>
achieved 89.75 %
<term>
#8201The new model achieved 89.75% F-measure, a 13% relative decrease in F-measure error over the baseline model’s score of 88.2%.
of 88.2 % . The article also introduces a
new
<term>
algorithm
</term>
for the
<term>
boosting
#8231The article also introduces a new algorithm for the boosting approach which takes advantage of the sparsity of the feature space in the parsing data.