AXON: A personalized retrieval information system in Arabic texts based on linguistic features

Abstract

The work presented in this paper aims to develop a system for information retrieval which has the particularity of being adapted to the Arabic language and providing personalized results based on the user's preferences / interests. For this reason, we proposed a query expansion method based on both semantic knowledge from existing ontologies for the Arabic language and information from the users' profiles. The proposed method is interesting for several reasons. Personalized information related to the users' profiles to enrich the query should be considered since it is entirely appropriate and justified to improve traditional IR systems. Moreover, this idea has been little exploited for IR systems in Arabic. In addition, the coupling between information from the user's profiles and that contained in the ontologies is also an original and interesting track to develop. The suggested query expansion method has been implemented and tested on 150 queries for 50 users or 7500 queries of evaluation. The results of this evaluation are promising.