A Comparison of Re-ranking Methods in Digital Libraries Using User Profiles

Abstract : In this paper we present a comparative study of re-ranking methods using user profiles in digital libraries which contain scientific articles. While most of present personalized search systems only use content-based methods to re-rank search results, in our work we use many different content-based and citation-based methods for this purpose. We also study many functions to combine scores computed by these methods. We conducted experiments on the test collection used in the INEX 2005 campaign to evaluate our strategies and received interesting results.
Type de document :
Communication dans un congrès
International Conference on Web Intelligence and Intelligent Agent Technology, 2008. WI-IAT '08. IEEE/WIC/ACM, Dec 2008, Sydney, Australia. 2008, 〈10.1109/WIIAT.2008.206〉
Liste complète des métadonnées

https://hal-emse.ccsd.cnrs.fr/emse-00674964
Contributeur : Florent Breuil <>
Soumis le : mardi 28 février 2012 - 15:46:03
Dernière modification le : mardi 23 octobre 2018 - 14:36:08

Identifiants

Citation

Thanh Trung Van, Michel Beigbeder. A Comparison of Re-ranking Methods in Digital Libraries Using User Profiles. International Conference on Web Intelligence and Intelligent Agent Technology, 2008. WI-IAT '08. IEEE/WIC/ACM, Dec 2008, Sydney, Australia. 2008, 〈10.1109/WIIAT.2008.206〉. 〈emse-00674964〉

Partager

Métriques

Consultations de la notice

87