An Integrated Approach for Context-Aware Query Recommendation in Folksonomies

Collaborative tagging systems have recently emerged as one of the most popular tools for Web search users to find their desired information. Indeed, the informal social classification structure in these systems, also known as folksonomy, provides a convenient way to annotate resources by allowing users to use any keyword or tag that they find relevant. In turn, the flat and non-hierarchical structure with unsupervised vocabularies leads to low search precision and poor resource navigation and retrieval. The widely keyword-based approaches used for locating information on the Web, are hence not straightforwardly adaptable to folksonomies. This drawback has created the need for an effective framework to support folksonomy users in effectively retrieving the resources matching their real search intents. The primary focus of this paper is to propose an integrated approach for context-aware query recommendation in folksonomies that exploits the power of both Hidden Markov Models (HMMs) and triadic concepts. We demonstrate through carried out experiments that the proposed approach yields mainly highly Precise and highly Relevant tag query recommendation MOTS-CLÉS : MMC, folksonomie, intention de recherche, contexte, concept triadique