#392At MIT Lincoln Laboratory, we have been developing a Korean-to-English machine translation system CCLINC (Common Coalition Language System at Lincoln Laboratory).
automated acquisition of grammars
</term>
.
Having
been trained on
<term>
Korean newspaper articles
#515(iii) Rapid system development and porting to new domains via knowledge-based automated acquisition of grammars. Having been trained on Korean newspaper articles on missiles and chemical biological warfare, the system produces the translation output sufficient for content understanding of the original document.
</term>
and
<term>
information sources
</term>
. We
have
built and will demonstrate an application
#815We have built and will demonstrate an application of this approach called LCS-Marine.
when a
<term>
request
</term>
is complete . We
have
demonstrated this capability in several
#888We have demonstrated this capability in several field exercises with the Marines and are currently developing applications of this technology in new domains.
Automatic Speech Recognition technology
</term>
have
put the goal of naturally sounding
<term>
#918Recent advances in Automatic Speech Recognition technologyhave put the goal of naturally sounding dialog systems within reach.
improved
<term>
speech recognition
</term>
has
brought to light a new problem : as
<term>
#936However, the improved speech recognitionhas 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.
system response
</term>
to
<term>
users
</term>
has
been extensively studied by the
<term>
natural
#975The issue of system response to usershas been extensively studied by the natural language generation community, though rarely in the context of dialog systems.
</term>
, where each
<term>
word string
</term>
has
been obtained by using a different
<term>
#1106The oracle knows the reference word string and selects the word string with the best performance (typically, word or semantic error rate) from a list of word strings, where each word stringhas been obtained by using a different LM.
concatenation grammar [ RCG ] formalism
</term>
has
revealed many attractive properties which
#1609The theoretical study of the range concatenation grammar [RCG] formalismhas revealed many attractive properties which may be used in NLP.
a
<term>
natural language generator
</term>
have
recently been proposed , but a fundamental
#2023Techniques for automatically training modules of a natural language generatorhave recently been proposed, but a fundamental concern is whether the quality of utterances produced with trainable components can compete with hand-crafted template-based or rule-based approaches.
word-level alignment models
</term>
does not
have
a strong impact on performance . Learning
#2649Surprisingly, learning phrases longer than three words and learning phrases from high-accuracy word-level alignment models does not have a strong impact on performance.
for
<term>
language understanding
</term>
and
have
a high
<term>
accuracy
</term>
but little robustness
#3525FSM provides two strategies for language understanding and have a high accuracy but little robustness and flexibility.
shortcomings of different strategies and
has
all the advantages of the three strategies
#3564Cooperative Model incorporates all the three strategies together and thus can suppress all the shortcomings of different strategies and has all the advantages of the three strategies.
baseline methods
</term>
. Previous research
has
demonstrated the utility of
<term>
clustering
#3887Previous research has demonstrated the utility of clustering in inducing semantic verb classes from undisambiguated corpus data.
system
</term>
.
<term>
Link detection
</term>
has
been regarded as a core technology for
#4045Link detectionhas been regarded as a core technology for the Topic Detection and Tracking tasks of new event detection.
understanding process
</term>
. Experiment results
have
shown that a
<term>
system
</term>
that exploits
#4258Experiment results have shown that a system that exploits the proposed method performs sufficiently and that holding multiple candidates for understanding results is effective.
in Kyoto city bus information system that
has
been developed at our laboratory . Experimental
#4398Dialogue strategies based on the user modeling are implemented in Kyoto city bus information system that has been developed at our laboratory.
that
<term>
manually sense-tagged data
</term>
have
in their sense coverage . Our analysis
#4911On a subset of the most difficult SENSEVAL-2 nouns, the accuracy difference between the two approaches is only 14.0%, and the difference could narrow further to 6.5% if we disregard the advantage that manually sense-tagged datahave in their sense coverage.
probabilistic translation models
</term>
that
have
recently been adopted in the literature
#5690This paper investigates some computational problems associated with probabilistic translation models that have recently been adopted in the literature on machine translation.
time-sensitive ( e.g . How many victims
have
been found ? ) Judges found
<term>
sentences
#5802Because of the dynamic nature of the stories, many questions are time-sensitive (e.g. How many victims have been found?)