Social bookmarking and tagging has emerged a new era in user collaboration.Collaborative Tagging allows users to annotate content of their liking, whichvia the appropriate algorithms can render useful for the provision of productrecommendations. It is the case today for tag-based algorithms to workcomplementary to rating-based recommendation mechanisms to predict the userliking to various products. In this paper we propose an alternative algorithmfor computing personalized recommendations of products, that uses exclusivelythe tags provided by the users. Our approach is based on the idea of using thesemantic similarity of the user-provided tags for clustering them into groupsof similar meaning. Afterwards, some measurable characteristics of users'Annotation Competency are combined with other metrics, such as user similarity,for computing predictions. The evaluation on data used from a real-worldcollaborative tagging system, citeUlike, confirmed that our approachoutperforms the baseline Vector Space model, as well as other state of the artalgorithms, predicting the user liking more accurately.
展开▼
