Interfacing thought: cognitive aspects of human-computer interaction
Predicting category accesses for a user in a structured information space
SIGIR '02 Proceedings of the 25th annual international ACM SIGIR conference on Research and development in information retrieval
Communications of the ACM
ACM Computing Surveys (CSUR)
Empirical Evaluation of User Models and User-Adapted Systems
User Modeling and User-Adapted Interaction
Evolutionary Reinforcement of User Models in an Adaptive Search Engine
WI '03 Proceedings of the 2003 IEEE/WIC International Conference on Web Intelligence
Rule-based query personalization in digital libraries
International Journal on Digital Libraries
Personalized Search Based on User Search Histories
WI '05 Proceedings of the 2005 IEEE/WIC/ACM International Conference on Web Intelligence
Adaptive information retrieval system applied to digital libraries
WebMedia '06 Proceedings of the 12th Brazilian Symposium on Multimedia and the web
Contextual search using ontology-based user profiles
Large Scale Semantic Access to Content (Text, Image, Video, and Sound)
A collaborative filtering based re-ranking strategy for search in digital libraries
ICADL'05 Proceedings of the 8th international conference on Asian Digital Libraries: implementing strategies and sharing experiences
Hi-index | 0.00 |
Suitable ranking of results is crucial when searching large collections available in Digital Libraries (BDs) so the first document presented to users are the most relevant to a search criterion. Several works acknowledge the need to consider the user's preferences for ranking search results, but it is hard to determine the user profile model and how much weight individual preferences should have in each application. This paper proposes a technique for re-ordering the results of searches on BDs that combines the relevance of each result item for the posed query with the individual user's preferences. In this proposal, the user profile maintains the characteristics of the documents that he/she prefers. This profile is built dynamically and transparently, based on the values of some metadata elements describing the documents accessed by the user. The proposal was validated in a BD providing access to a collection of literature documents.