#4604We approximateArabic 's rich morphology by a model that a word consists of a sequence of morphemes in the pattern prefix*-stem-suffix* (* denotes zero or more occurrences of a morpheme).
other,20-1-P03-1051,ak
sequence of
<term>
morphemes
</term>
in the
<term>
pattern
</term>
prefix * - stem-suffix * ( * denotes
#4622We approximate Arabic's rich morphology by a model that a word consists of a sequence of morphemes in thepattern prefix*-stem-suffix* (* denotes zero or more occurrences of a morpheme).
other,30-7-P03-1051,ak
manually segmented corpus
</term>
of the
<term>
language
</term>
of interest . A central problem of
#4797We believe this is a state-of-the-art performance and the algorithm can be used for many highly inflected languages provided that one can create a small manually segmented corpus of thelanguage of interest.
other,32-5-P03-1051,ak
parameters
</term>
with the expanded
<term>
vocabulary
</term>
and
<term>
training corpus
</term>
.
#4740To improve the segmentation accuracy, we use an unsupervised algorithm for automatically acquiring new stems from a 155 million word unsegmented corpus, and re-estimate the model parameters with the expandedvocabulary and training corpus.
other,35-1-P03-1051,ak
denotes zero or more occurrences of a
<term>
morpheme
</term>
) . Our method is seeded by a
<term>
#4637We approximate Arabic's rich morphology by a model that a word consists of a sequence of morphemes in the pattern prefix*-stem-suffix* (* denotes zero or more occurrences of amorpheme).
tech,1-3-P03-1051,ak
unsegmented Arabic corpus
</term>
. The
<term>
algorithm
</term>
uses a
<term>
trigram language model
#4673Thealgorithm uses a trigram language model to determine the most probable morpheme sequence for a given input.
tech,17-2-P03-1051,ak
</term>
and uses it to bootstrap an
<term>
unsupervised algorithm
</term>
to build the
<term>
Arabic word segmenter
#4657Our method is seeded by a small manually segmented Arabic corpus and uses it to bootstrap anunsupervised algorithm to build the Arabic word segmenter from a large unsegmented Arabic corpus.
tech,2-6-P03-1051,ak
training corpus
</term>
. The resulting
<term>
Arabic word segmentation system
</term>
achieves around 97 %
<term>
exact match
#4747The resultingArabic word segmentation system achieves around 97% exact match accuracy on a test corpus containing 28,449 word tokens.
tech,22-2-P03-1051,ak
unsupervised algorithm
</term>
to build the
<term>
Arabic word segmenter
</term>
from a
<term>
large unsegmented Arabic
#4662Our method is seeded by a small manually segmented Arabic corpus and uses it to bootstrap an unsupervised algorithm to build theArabic word segmenter from a large unsegmented Arabic corpus.
tech,9-5-P03-1051,ak
segmentation accuracy
</term>
, we use an
<term>
unsupervised algorithm
</term>
for automatically acquiring new
<term>
#4717To improve the segmentation accuracy, we use anunsupervised algorithm for automatically acquiring new stems from a 155 million word unsegmented corpus, and re-estimate the model parameters with the expanded vocabulary and training corpus.
tech,9-7-P03-1051,ak
state-of-the-art performance and the
<term>
algorithm
</term>
can be used for many
<term>
highly
#4776We believe this is a state-of-the-art performance and thealgorithm can be used for many highly inflected languages provided that one can create a small manually segmented corpus of the language of interest.