A00-2020 |
weighted sum uses a pseudo-count
|
predictor
|
. This predictor computes the
|
A00-2017 |
experiment uses MLE , the majority
|
predictor
|
. In addition , we conducted
|
A00-2017 |
to the performance of the word
|
predictor
|
. Finally , we describe a large
|
A00-2017 |
The application in which a word
|
predictor
|
is used might give a partial
|
A00-2003 |
by a linear combination of the
|
predictor
|
variables . Variable weights
|
A00-2020 |
every context . In fact , these
|
predictors
|
can be any probability distribution
|
A00-2028 |
Automatic Problematic Dialogue
|
Predictor
|
Our experiments apply the machine
|
A00-1021 |
using a linear function of seven
|
predictor
|
variables . We provide an evaluation
|
A00-2017 |
POS tags in order to derive the
|
predictor
|
. In this paper we show that
|
A00-2029 |
looks at prosody as one possible
|
predictor
|
of ASR performance . ASR performance
|
A00-2005 |
the value predicted by the most
|
predictors
|
, the majority vote . 2.2 Bagging
|
A00-2020 |
pseudo-count predictor . This
|
predictor
|
computes the probability of an
|
A00-2028 |
generalize our problematic dialogue
|
predictor
|
to other systems . Thus we also
|
A00-2028 |
automatic problematic dialogue
|
predictor
|
( Cohen , 1996 ) . Section 4
|
A00-2028 |
likely to produce generalized
|
predictors
|
( Litman et al. , 1999 ) . The
|
A88-1002 |
words and phrases that were good
|
predictors
|
of a particular topic but occurred
|
A00-2005 |
training set . Algorithm : Bagging
|
Predictors
|
( Breiman , 1996 ) ( 1 ) Given
|
A00-2028 |
input features that are used as
|
predictors
|
for the classes . We start with
|
A00-2017 |
to be used when evaluating the
|
predictors
|
. Tables 4,5 present the results
|
A00-1031 |
- ing . The suffix is a strong
|
predictor
|
for word classes , e.g. , words
|