lr-prod,11-3-P99-1080,ak |
to be tested on a standard task ,
<term>
|
The Wall Street Journal
|
</term>
, allowing a fair comparison with
|
#31219
The model is to be tested on a standard task,The Wall Street Journal, allowing a fair comparison with the well-known tri-gram model. |
|
tech,4-1-P99-1080,ak |
language
</term>
. This paper discusses a
<term>
|
decision-tree approach
|
</term>
to the problem of assigning
<term>
|
#31174
This paper discusses adecision-tree approach to the problem of assigning probabilities to words following a given text. |
|
tech,10-2-P99-1080,ak |
language model attempts
</term>
, an
<term>
|
algorithm
|
</term>
for selecting
<term>
nearly optimal
|
#31199
In contrast with previous decision-tree language model attempts, analgorithm for selecting nearly optimal questions is considered. |
|
measure(ment),11-1-P99-1080,ak |
</term>
to the problem of assigning
<term>
|
probabilities
|
</term>
to
<term>
words
</term>
following a given
|
#31181
This paper discusses a decision-tree approach to the problem of assigningprobabilities to words following a given text. |
|
other,17-1-P99-1080,ak |
<term>
words
</term>
following a given
<term>
|
text
|
</term>
. In contrast with previous
<term>
|
#31187
This paper discusses a decision-tree approach to the problem of assigning probabilities to words following a giventext. |
|
model,4-2-P99-1080,ak |
</term>
. In contrast with previous
<term>
|
decision-tree language model attempts
|
</term>
, an
<term>
algorithm
</term>
for selecting
|
#31193
In contrast with previousdecision-tree language model attempts, an algorithm for selecting nearly optimal questions is considered. |
|
other,13-2-P99-1080,ak |
<term>
algorithm
</term>
for selecting
<term>
|
nearly optimal questions
|
</term>
is considered . The
<term>
model
</term>
|
#31202
In contrast with previous decision-tree language model attempts, an algorithm for selectingnearly optimal questions is considered. |
|
model,1-3-P99-1080,ak |
questions
</term>
is considered . The
<term>
|
model
|
</term>
is to be tested on a standard task
|
#31209
Themodel is to be tested on a standard task, The Wall Street Journal, allowing a fair comparison with the well-known tri-gram model. |
|
other,13-1-P99-1080,ak |
assigning
<term>
probabilities
</term>
to
<term>
|
words
|
</term>
following a given
<term>
text
</term>
|
#31183
This paper discusses a decision-tree approach to the problem of assigning probabilities towords following a given text. |
|
model,23-3-P99-1080,ak |
fair comparison with the well-known
<term>
|
tri-gram model
|
</term>
.
|
#31231
The model is to be tested on a standard task, The Wall Street Journal, allowing a fair comparison with the well-knowntri-gram model. |
|