...
www.cs.cmu.edu/tom/newchapters.html.
[22] T. S. Jayram, Rajasekar Krishna-murthy, Sriram Raghavan, Shivakumar
Vaithyanathan, and Huaiyu Zhu. Avatar information extraction system. IEEE Data
Eng. Bull. ...
learning framework, which overcomes the challenges about scalability and adaptability
of the previous approaches.
We have then adapted the probabilistic framework to a Vietnamese domain - real ...
based. We also adapt the probabilistic framework to Vietnamese Real Estate domain
and have a satisfactory result.
1.4 Chapter summary
This chapter brought an overview of web-page problem and...
... and class formation, but rather as a
frameworkfor de®ning an agenda of problems for empirical research within
class analysis. In the multivariate empirical studies of class conscious-
ness and ... framework
10. A general frameworkfor studying
class consciousness and class
formation
In one way or another, most class analysts believe that at the core of class
analysis is a relatively simple causal ... of analysis, typically an entire society. To say that someone is
``in'' a managerial class location is to claim that they are embedded in a
set of interindividual interactions (relations)...
... disadvantage noncompliance. Law is also
similar to what Wiener and Doescher (1991)'term a structural
solution, that is, a political act that mandates individual
behavior. For Taylor and ... have changed dramati-
cally in the past years, and as a result, policy with respect to
managing tobacco usage behavior also has changed. The re-
lationship of behavior management and externalities ... singular choice of edu
cation, marketing, or force of law as a paradigm of choice.
Each paradigm has a role to play in behavior management;
behavior management must be considered from the
pragmatic...
... system
learns this as a non-transliteration but it is wrongly
annotated as a transliteration in the gold standard.
Arabic nouns have an article “al” attached to them
which is translated in English as ... International
Language Resources and Evaluation (LREC’10), Val-
letta, Malta.
Sittichai Jiampojamarn, Kenneth Dwyer, Shane Bergsma,
Aditya Bhargava, Qing Dou, Mi-Young Kim, and
Grzegorz Kondrak. ... non-transliterations by N.
3.2 Implementation Details
We use the Forward-Backward algorithm to estimate
the counts of multigrams. The algorithm has a for-
ward variable α and a backward variable...
... that are assigned the
correct head. Complete accuracy (CA) indicates
the fraction of sentences that have a complete cor-
rect analysis. We also measure that root accuracy
(RA) and leaf accuracy ... words
after w
2
(as in (Yamada and Matsumoto, 2003)).
The key additional feature we use, relative to (Ya-
mada and Matsumoto, 2003), is that we include
the previous predicted action as a feature. ... more
conjunctive features. Overall, the average number
of active features in an example is about 50.
4.2 Evaluation
We use the same evaluation metrics as in (McDon-
ald et al., 2005). Dependency accuracy (DA)...
... for Computational Linguistics
Archivus: Amultimodal system for multimedia meeting browsing and
retrieval
Marita Ailomaa, Miroslav Melichar,
Martin Rajman
Artificial Intelligence Laboratory
´
Ecole ... certain modalities has an im-
pact on how they choose to use language when all
modalities are available.
On the backend, the wizards can also to some
extent have an active role in encouraging language
use. ... controlled
for in the experiment increases substantially. For
instance, if it is the case that within a single inter-
face any task that can be performed using natural
language can also be performed...
... System architecture
The underlying database of movie information is
stored in XML format. When a new database is
available, a Grammar Compiler component ex-
tracts and normalizes the relevant ... from the
database. These are used in conjunction with a pre-
defined multimodal grammar template and any
available corpus training data to build a multimo-
dal understanding model and speech ... scenarios: tablet versus remote
control. Our main goals for the data collection and
statistical analysis were three-fold: collect a large
corpus of natural multimodal dialogue for this me-
dia selection...
...
evaluation metrics are able to closely approximate
human evaluations for various applications. Given
an application app and an evaluation guideline
package eval, the faithfulness/compactness ...
separately evaluated. Each version was evaluated
by a human evaluator, with no reference answer
available. For this evaluation 115 test questions
were used, and the human evaluator was asked ... predicts that the automatic
evaluation metrics that explain most of the
variation in the human evaluation must have a
median α and a median N. As seen in Table 6, our
evaluation framework correctly...
... states that repairs are
available for semantic analysis but provides
no details on the representation to be used.
Clearly repairs should be available for se-
mantic analysis as they play a ...
work has proved adequate fora collection of
human-human task-oriented dialogs, both in
a full manual examination of the corpus, and
in tests with a parser capable of parsing some
of that corpus. ...
To test this framework, data was examined
from 31 TRAINS 93 dialogs (Heeman and
Allen, 1995), a series of human-human prob-
lem solving dialogs in a railway transporta-
tion domain. 5 There...
... representation language from the applica-
tion program and systematically translating it into
a natural language using syntactic/semantic rules
which were primarily designed for translating in
the ... (1) must include at
least as much information as we have given above.
In particular, the logical structure of our para-
phrases contains essential information (about, for
instance, the differences ... representa-
tion of linguistic knowledge for both analysis and
generation. We argue that the only part of the
average NL system's knowledge that we can have
any faith in is its vocabulary and,...
... lexical sample format, which is an
XML–based format that has been used for both the
S
ENSEVAL-2 and SENSEVAL-3 exercises. A file in
this format includes a number of instances, each one
made up ... Poster and Demonstration Sessions,
pages 73–76, Ann Arbor, June 2005.
c
2005 Association for Computational Linguistics
SenseRelate::TargetWord – A Generalized Framework
for Word Sense Disambiguation
Siddharth ... Disambiguation, the ar-
chitecture and usage of SenseRelate::TargetWord,
and a description of the user interfaces (command
line and GUI).
2 The Framework
The package has a highly modular architecture....
... to
re-use language resources, and that standards for
linguistic annotation are becoming increasingly
mandatory. In particular, there is a need for a
general frameworkfor linguistic annotation that
is ... existing annotated data
in XML. Tools for editing annotations in the
abstract format, which automatically generate
virtual AML from Data Category and Dialect
Specifications, are already under development
in ... of the annotation framework: it defines a
lingua franca for syntactic annotations that can
be used to compare and merge annotations, as
well as enable design of generic tools for
visualization,...
...
4 Information Structure and
Tactical Constraints
Entries in the eategorial lexicon have tactical con-
straints, grammatical and semantic features, and
phonological representation. Similar ... Multi-dimensional compo-
sitional functions as a basis for grammatical anal-
ysis. In R. T. Oehrle, E. Bach, and D. Wheeler,
editors, Categorial Grammars and Natural Lan-
guage Structures, D. Reidel, ...
INT-clear situation
'Very clear situation'
uzun
yol
long road
'long
road'
bu- ndan ba~ka
this-ABLATIVE other
'other than this'
ktz kedi-yi gSr-dii
girl cat-ACC...
... information absent
in surface utterances such as zero anaphora are
compensated for and in some cases multiple
analyses are disambiguated. (3-3), the surface
analysis of (3-2), is analyzed as ... diagrams an overview of the
procedure for translating speaker's meaning. In
contrast to a conventional machine translation
procedure, speaker's meaning can be analyzed
and generated, ... 180, JAPAN
Abstract
A frameworkfor translating speaker's
meaning or intention is proposed based on two
notions, Illocutionary Force Types (IFTs) for
analysis and Decision Parameters...
... not matching the available references are
evaluated as poor or bad translations.
The main objective of this work is to propose
and evaluate AM-FM, a semantic frameworkfor
assessing translation ... consistent, fast and cheap, automatic
evaluation has the major disadvantage of requiring
reference translations. This makes automatic eval-
uation not reliable in the sense that good transla-
tions ... evaluation dataset had to be used.
156
2 Related Work and Dataset
Although BLEU (Papineni et al., 2002) has be-
come a de facto standard for machine translation
evaluation, other metrics such as...