Social tagging is an information classification paradigm where the users themselves are given the power to describe and categorize content for their own purposes using tags. The popularity of social tagging, and the ease with which such tags can be generated, assigned, and collected, has sparked significant research interest in tags and their possible applications. One such application is tag-based recommendation: generating better recommendations by incorporating tags into the recommendation process. This chapter provides an overview of the state-of-the-art approaches to tag-based item recommendation, organised by the class of recommendation algorithms that is augmented with tags, such as collaborative filtering, dimensionality reduction, graph-based recommendation, content-based filtering, machine learning, and hybrid recommendation. The chapter also offers an overview of the most important methods for recommending which tags to apply to content. Finally, the chapter discusses the open research problems in tag-based recommendation and what would be needed to address them.
|Titel||Social Information Access|
|Redaktører||Peter Brusilovsky, Daqing He|
|Publikationsdato||1 jan. 2018|
|Status||Udgivet - 1 jan. 2018|
|Navn||Lecture Notes in Computer Science|