Personalized and context-aware retrieval based on fuzzy ontology profiling

Web Information retrieval aims to satisfy users' search needs which are highly dependent on their interests and preferences. In this context, web personalization provides an adapted information retrieval. Specific computational modeling is required to address the large amount of heterogeneous electronic documents and the lack of semantic annotations on the web which makes knowledge discovery challenging. In this paper, a novel system, based on fuzzy ontological user profile is proposed. The latter, is composed of history, positive and negative preferences. An implicit relevance judgments method is also introduced. Furthermore, the system is context-aware by integrating novel contextual similarity measures and supporting semantic fuzzification. Our proposal has been implemented and has endured a twofold evaluation. The results show that the proposed system can provide more personalized results and confirm the interest of the context-aware search.