HMM-CARe: Hidden Markov models for context-aware tag recommendation in folksonomies

Collaborative tagging systems allow users to manually annotate web resources with freely chosen keywords aka tags without any restriction to a certain vocabulary. The resulting collection of all these users annotations constitute the so-called folksonomy. Such systems typically provide simple tag recommendations skills to increase the number of tags assigned to resources. In this this paper, we propose a novel Hidden Markov Model (HMM) based approach, called HMM-CARE, for tags recommendation. Specifically, we extend the HMM to include user's tagging intents, formally represented as triadic concepts. Carried out experiments emphasize the relevance of our proposal and open many thriving issues.