... training data
are comparatively smaller than WAQ and WAQA,
they however yield comparable results. The linear
combination of datasets (WAQ+WAQA+LSR
Lin
)
yields statistically significant performance ... statistical word as-
sociations which are trained on parallel monolin-
gual corpora. The major drawback of this ap-
proach lies in the limited availability of truly par-
allel monolingual corpor...
... channel
was down-sampled to 8kHz and segmented using an
available broadcast news segmenter. Because we did
not have a pronunciation dictionary which covered
the transcribed audio, we automatically ... does assume that they belong to an in-
terval scale. Similarly, the arithmetic mean of MAP
assumes AP has interval scale. As Robertson (2006)
has pointed out, it is in no sense clear that AP
(...
... Modifiers with a Reranking Approach
Jenny Liu
MIT CSAIL
jyliu@csail.mit.edu
Aria Haghighi
MIT CSAIL
me@aria42.com
Abstract
In this work, we present a novel approach
to the generation task of ordering ... We can
express this as a real-valued feature:
φ(B,H, x)=
count in training data of all
n-grams present in x
See Table 2 for a summary of our features. Many
of the features we use a...
... the data. If a word ap-
pears at least K times in the data, the supertagger
only considers categories that appear in the word’s
category set, rather than all lexical categories.
The second parsing ... Adjoining Grammar
as an alternative to context-free grammar, and
here we use another “mildly context-sensitive” for-
malism, Combinatory Categorial Grammar (CCG,
Steedman (2000)), which arguab...
... grammar. An exam-
ples is:
S ~ al S S~al A1
S-+anS S-+anAn
A~ -+ a~ X
A2 + al A2
An -~ al An
X-+e
A1 -+ a2 Az A1 ~ an A1
A2 -+ a2 X A2 ~ an A2
An -+ a2 A, ~ An ~ an X
Here the grammar ... context-free grammars because context-free lan-
guages are the smallest class of formal language that
can realistically be applied to the analysis of natural
language. Techniques suc...
... param-
eters associated with an event z, φ a notation for
all model parameters, and X a notation for all ran-
dom variables that represent observable features.
2
Given a document collection annotated ... dia of Philosophy (Fall 2009
Edition), Edward N. Zalta (ed.), http://plato.stan
ford.edu/archives/fall2009/entries/davidson/.
Srini Narayanan and Sanda Harabagiu. 2004. Ques-
tion Ans...
... unlabeled data
and its automatic Chinese translation, and vice
versa.
Although not as significant as those with parallel
data, we can still obtain improvements using the
pseudo-parallel data, ... Banea, Rada Mihalcea, and Janyce Wiebe.
2010. Multilingual subjectivity: Are more languages
better? In Proceedings of COLING’10.
Carmen Banea, Rada Mihalcea, Janyce Wiebe, and
Samer Hass...
... forward and backward
Viterbi algorithm, which is almost the same as
calculating Eq. 3 with a variant of the forward-
backward algorithm (Sha and Pereira, 2003). The
same numerical optimization ... non-linear measures such as F-
score, while all of the above criteria achieve op-
timization based on the linear combination of av-
erage accuracies, or error rates, rather than a given
task-s...
... reason is
that f
3
may appears in various phrases, such as
“
, accept France ’s invitation”.
While f
2
almost always appears in f
1
, indicating
that the variable X may not be replaced with other
words ... C-value, a measurement
of automatic term recognition, to score source
phrases. A source phrase is regarded as a key
phrase if its score greater than a threshold. Note
that a sou...
... evaluation, on the one
hand, can be .carried out automatically in a large
scale, on the other hand, can suggest what the
direct evaluation entails in some way because that
none appropriate ...
their accuracy rates and loss rates manually. Tab.
5 lists the results.
Ta~ num.
0
2
Aver. accuracy(%) Aver. loss(%)
94.6 7.3
90.1 5.2
87.6 2.1
Tab. 5. Average accuracy and loss rates .....