Research Reports
Permanent URI for this collectionhttp://dl.cerist.dz/handle/CERIST/34
Browse
2 results
Search Results
Item Matching Resources in Social Environment(CERIST, 2012-05) Benna, AmelUser comments on web become more and more important. We focus, in this paper, on the use of user-defined tags for annotating resources to identify links between them. These links are based on a resource social context, obtained by applying k-means classification method and a hierarchical classification of tags within a cluster. The resources are re-assigned to this classification, to facilitate search process, and the ranking of result is performed according to their degree of relevance, by evaluating a similarity score between the tagged contents, in hierarchical clusters of tags, and the user request. The results of the evaluation, on social bookmarking system del.icio.us, demonstrate significant improvements over traditional approaches.Item Building a social network, based on collaborative tagging, to enhance social information retrieval(CERIST, 2012) Benna, Amel; Mellah, HakimaWeb 2.0 technologies put user at the center of data production and introduce a strong social collaboration. Therefore, the techniques used in traditional information retrieval systems do not meet the requirements of users who want to take into account their social preferences. The idea reported in this paper is to include not only the social context of the user but also that of the resource. In the social network that we consider, the user social context brings his interests, which are captured from a collaborative tagging system, while the resource social context is related to clusters of tags, obtained by classification method, and users’ opinions according to their expertise level on the resource. The results of our experiment evaluation on real-world dataset (crawled from delicious folksonomy) demonstrate significant improvements over traditional retrieval approaches.