An Ontology-based Personalized Retrieval Model Using Case Base Reasoning.

Amir Zidi (1) , Amna Bouhana (2) , Mourad Abed (1) , Afef Fekih (3)

1 LAMIH - Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201
2 Université de Sfax - University of Sfax
3 University of Louisiana at Lafayette

A novel ontology Based Personalized Retrieval model using the Case Base Reasoning (CBR) tool is designed and presented in this paper. The proposed approach is aimed at achieving a scalable and user friendly data retrieval system with high retrieval performance where search results are ranked based on user preferences. The proposed retrieval framework integrates the advantages of two methods, a content based method (ontology) to represent data and a case based method (CBR) to personalize the search process and to provide users with alternative documents recommendations. To analyze the performance of the proposed approach, computer experiments are carried out using recall precision curve and average precision (AP) metric. The performance of our approach is then compared to a framework that uses the classic vector space model. Results clearly indicate the strength of the proposed appr oach as well as its ability to accurately retrieve pertinent information. The proposed approach is particularly promising in applicable related to city logistics, especially in the field of itinerary research for urban freight transport.

Mots clés: "Case base reasoning ontology"; Personalized search; Information retrieval; Query formulation

Domaines: Informatique [cs]

Amir Zidi, Amna Bouhana, Mourad Abed, Afef Fekih. An Ontology-based Personalized Retrieval Model Using Case Base Reasoning.. 18th International Conference in Knowledge Based and Intelligent Information and Engineering Systems, KES 2014, Sep 2014, Gdynia, Poland. pp.213-222, ⟨10.1016/j.procs.2014.08.101⟩. ⟨hal-03387811⟩