For both stemmer and root extraction Alkhalil. It was implemented using Java. However, it has many outputs with different patterns. So, you want to choose from these outputs the most important or suitable one. It is a comprehensive stemmer. You may use it natural language understanding. If you are going to use the stemmer in information retrieval I suggest Khoja algorithms because it unambiguous and easy to implement and in IR domain the errors in stemmer maybe accepted.

thanks Dear Mohammed i will look over khoja algo and as  you said for alkhalil stemmer is comprehensive.

There is a survey on Arabic stemmers

http://www.abdelali.net/ref/AlSughaiyer_ArabicMorphoTechniques_04.pdf

I think Alkhalil is more recent than this survey. I hope this survey is useful 

Is Alkhalil a free tool or we have to re-implement i?

I think also you have to see the  Nizar Habash homepage, you can found free tools.

Dear Mohammed i did read the article and its old. Dear karimah thank ,, and i  will  look at nizar habash website.

Alkhalil is free tool. let me search and come back to you with the link

i think khoja algo

You can download Alkhalil from this link

http://sourceforge.net/projects/alkhalil/

I hope it will help

Also, if you want to read about Alkhalil in Arabic, kindly see this link

http://itwadi.com/node/2204

Dear Bassam ;

Please, check these packages:

http://lucene.apache.org/

https://gate.ac.uk/

They include libraries for Arabic collections. Also, Stanford University has some packages about Arabic and sentiment analysis

i am using lucene right now thanks osaman

In 2008 and 2009 we organised two tracks at the Cross-Language Evaluation Forum (CLEF) about Persian Information Retrieval.

You can find all the info about the experiments carried out by participants in the Publication area (working notes and papers) of the following link:

http://www.clef-initiative.eu/home

I'm sure you can find something useful for your stemming problem.

Dear Bassam,

How did you find Lucene? In which domain will you use it?

Dear Mohammed

i am using Lucene right now.. i find it diffcult  to use i need much time  to understand every thing about it such as lexical analysis and stemming and removing stop words,, i am traying  to build an Arabic text calssifier

Dear Bassam,

you can check this

https://lucene.apache.org/core/2_9_4/api/contrib-analyzers/org/apache/lucene/analysis/ar/ArabicLetterTokenizer.html

https://lucene.apache.org/core/2_9_4/api/contrib-analyzers/org/apache/lucene/analysis/ar/ArabicStemmer.html

https://lucene.apache.org/core/2_9_4/api/contrib-analyzers/org/apache/lucene/analysis/ar/ArabicStemFilter.html

https://lucene.apache.org/core/2_9_4/api/contrib-analyzers/org/apache/lucene/analysis/ar/ArabicNormalizer.html

https://lucene.apache.org/core/2_9_4/api/contrib-analyzers/org/apache/lucene/analysis/ar/ArabicNormalizationFilter.html

Also, you can use Lucene in Action book to know the basic in English for using Lucene library. Then you can build your Arabic classifier. Also, I think there is classifier libraries (source codes) if you google about it.

Also, you can use www.github.com website and http://sourceforge.net/ among others for available source code.

Also, check this survey by now publisher for Arabic IR

http://www.nowpublishers.com/article/Details/INR-031

thanks dear osman i am reading now about the library and also used source code from http://sourceforge.net/ which i find it very helpfull

Please have a look at the below link. "The ISRI Arabic Stemmer"

http://www.nltk.org/_modules/nltk/stem/isri.html

NTLK and some other libraries contains Arabic stemmer but I think the fastest IR package is Lucene. Also, you check GATE library for information Extraction https://gate.ac.uk/gate/doc/plugins.html#Lang_Arabic and it is especially working for Arabic ontology https://gate.ac.uk/sale/tao/splitch13.html#x18-33800013.7

More recent POS taggers: Farasa, ADAM, Multi-dialect Arabhic Pos tagging: A CRF approach and Part of Speech tagging for Arabic gulf Dialect using Bi-LSTM

The best for information retrieval in terms of MAP and P@ is Farasa stemmer compared to light stemming, Khodja, MADMAIRA, you can refer to Abdelali A, Darwish K, Durrani N et al. Farasa: A fast and furious segmenter for arabic. In Proceedings of the

Demonstrations Session, NAACL HLT 2016, The 2016 Conference of the North American Chapter of the Association

for Computational Linguistics: Human Language Technologies, San Diego California, USA, June 12-17, 2016. pp.

11–16.