Login New user?  
The Journal of Applied Science University
An International Journal


Volumes > Vol. 06 > No. 02


A Priority Based Approach for Stemming Arabic Word

PP: 377-380
Riyad Al-Shalabi,
Recently Natural Language Processing of Arabic text has gained a lot of attention. We need to understand many problems in Arabic linguistics in order to carry out the Natural Language Processing projects we want to attack. This paper describes an algorithm for stemming Arabic words. The algorithm calculates the root based on removing the suggested prefixes and suffixes and then finds out all the possibilities in which the root could be one of them, finally it starts minimizing these possibilities in order to find out the correct root based on the calculation of the number of the original letters of each possibility formed. If the word does not contain any original letters the algorithm tries to find the root letters according to some priorities assigned to letters as described below. We have tested our algorithm on a different set of Arabic abstracts taken from the proceedings of the Saudi Arabian National Computer Conferences, the performance of the algorithm is very high where in some abstracts we reached to 95%.

  Home   About us   News   Journals   Conferences Contact us Copyright naturalspublishing.com. All Rights Reserved