A collaborative filtering recommendation algorithm based on tag clustering

Abstract

Social tagging system is applied widely in Web 2.0 nowadays, which is designed to express the user's interest and willingness more accurately. And tag clustering is an important research topic in personalized recommendation of social tagging systems. This paper presents a personalized recommendation method based on tag clustering. In this method, tag clustering is realized by calculating the tag similarity, and recommendation is made based on tag clustering results. Experiments using CiteULike data sets show, proposed method can optimize ranking of objective resources, and help users to discover new resources easier. © 2014 Springer-Verlag.