Article

Construction des profils utilisateurs à base d'une ontologie pour une recherche d'information personnalisée.

01/2008;
Source: OAI

ABSTRACT La recherche d'information (RI) personnalisée tend principalement à modéliser l'utilisateur selon un profil puis à l'intégrer dans la chaîne d'accès à l'information, afin de mieux répondre à ses besoins spécifiques. Ce papier présente une extension d'une approche de construction implicite du profil utilisateur précédemment développée où les centres d'intérêts sont représentés à base de termes pondérés. L'extension de cette approche permet d'obtenir une représentation sémantique de ces centres à base de concepts pondérés en utilisant l'ontologie de l'ODP. Nous avons évalué notre approche sur la collection de documents TREC et avons présenté quelques résultats expérimentaux mettant en évidence l'impact de l'intégration du profil utilisateur sur la performance du système.

0 Followers
 · 
144 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: The current components' retrieval tools offer few services to facilitate a relevant search of these components. They are intended for experts having already a good knowledge of the components' catalogs. However, they are still limited for inexperienced users. In this context, the present contribution aims at analysing and comparing some existing retrieval approaches and studying the user profile representation. This study enabled us to incorporate the best practices of this field, to define our user profile and to carry out a process patterns' system guiding the integration of user profile in the search process.
    The 7th IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2009, Rabat, Morocco, May 10-13, 2009; 01/2009
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: today, the Web information is too heterogeneous that users have difficulties to retrieve their needed information: text, image or video. In this context, the collaborative work presents one solution proposed to solve this problem. Collaborative retrieval enables the retrieval histories' sharing between users having the same profile across multiple tools such as annotations. However the user is over whelmed by a variety of contradict annotations or even annotations which are far from the original subject. In this context, we propose an approach for the detection of relevant annotations to a given evidence source. The new concept of " relevant annotation " can be then, considered as a new source of information describing document used to improve collaborative retrieval performance. In our contribution, we propose a model of filtering of annotation given in semi-structured documents returned by collaborative retrieval systems. And for verify their relevance we propose using them for relevance feedback to extend the initial user's query.
    International Conference on Reasoning and Optimization in Information Systems: (ROIS’2013), Sousse, Tunisia;; 09/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: today, the Web information is too heterogeneous that users have difficulties to retrieve their needed information: text, image or video. In this context, the collaborative work presents one solution proposed to solve this problem. Collaborative retrieval enables the retrieval histories' sharing between users having the same profile across multiple tools such as annotations. However the user is over whelmed by a variety of contradict annotations or even annotations which are far from the original subject. In this context, we propose an approach for the detection of relevant annotations to a given evidence source. The new concept of " relevant annotation " can be then, considered as a new source of information describing document used to improve collaborative retrieval performance. In our contribution, we propose a model of filtering of annotation given in semi-structured documents returned by collaborative retrieval systems. And for verify their relevance we propose using them for relevance feedback to extend the initial user's query.
    International Conference on Reasoning and Optimization in Information Systems: (ROIS’2013), Sousse, Tunisia;; 09/2013

Preview

Download
18 Downloads
Available from