#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.
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.
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.
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.
tech,11-4-P03-1050,ak
can be used to further improve the
<term>
stemmer
</term>
by allowing it to adapt to a desired
#4499Monolingual, unannotated text can be used to further improve thestemmer by allowing it to adapt to a desired domain or genre.
model,20-6-P03-1050,ak
<term>
Arabic stemmer
</term>
built using
<term>
rules
</term>
,
<term>
affix lists
</term>
, and
<term>
#4554Our resource-frugal approach results in 87.5% agreement with a state of the art, proprietary Arabic stemmer built usingrules, affix lists, and human annotated text, in addition to an unsupervised component.
tech,3-7-P03-1050,ak
<term>
Task-based evaluation
</term>
using
<term>
Arabic information retrieval
</term>
indicates an improvement of 22-38
#4574Task-based evaluation usingArabic information retrieval indicates an improvement of 22-38% in average precision over unstemmed text, and 96% of the performance of the proprietary stemmer above.