#4599Task-based evaluation using Arabic information retrieval indicates an improvement of 22-38% in average precision over unstemmed text, and 96% of the performance of the proprietarystemmer above.
model,1-2-P03-1050,ak
non-English ( Arabic ) stemmer
</term>
. The
<term>
stemming model
</term>
is based on
<term>
statistical machine
#4449Thestemming model is based on statistical machine translation and it uses an English stemmer and a small (10K sentences) parallel corpus as its sole training resources.
tech,0-7-P03-1050,ak
<term>
unsupervised component
</term>
.
<term>
Task-based evaluation
</term>
using
<term>
Arabic information retrieval
#4571Our resource-frugal approach results in 87.5% agreement with a state of the art, proprietary Arabic stemmer built using rules, affix lists, and human annotated text, in addition to an unsupervised component.Task-based evaluation using Arabic information retrieval indicates an improvement of 22-38% in average precision over unstemmed text, and 96% of the performance of the proprietary stemmer above.
other,7-3-P03-1050,ak
parallel text
</term>
is needed after the
<term>
training phase
</term>
.
<term>
Monolingual , unannotated
#4485No parallel text is needed after thetraining phase.
lr,27-2-P03-1050,ak
parallel corpus
</term>
as its sole
<term>
training resources
</term>
. No
<term>
parallel text
</term>
is
#4475The stemming model is based on statistical machine translation and it uses an English stemmer and a small (10K sentences) parallel corpus as its soletraining resources.
other,16-7-P03-1050,ak
<term>
average precision
</term>
over
<term>
unstemmed text
</term>
, and 96 % of the performance of
#4587Task-based evaluation using Arabic information retrieval indicates an improvement of 22-38% in average precision overunstemmed text, and 96% of the performance of the proprietary stemmer above.
tech,34-6-P03-1050,ak
annotated text
</term>
, in addition to an
<term>
unsupervised component
</term>
.
<term>
Task-based evaluation
</term>
#4568Our resource-frugal approach results in 87.5% agreement with a state of the art, proprietary Arabic stemmer built using rules, affix lists, and human annotated text, in addition to anunsupervised component.
tech,4-1-P03-1050,ak
users
</term>
. This paper presents an
<term>
unsupervised learning approach
</term>
to building a
<term>
non-English (
#4436This paper presents anunsupervised learning approach to building a non-English (Arabic) stemmer.