Solving the Cold-Start Problem in Recommender Systems with Social Tags - Computer Science > Information RetrievalReport as inadecuate




Solving the Cold-Start Problem in Recommender Systems with Social Tags - Computer Science > Information Retrieval - Download this document for free, or read online. Document in PDF available to download.

Abstract: In this paper, based on the user-tag-object tripartite graphs, we propose arecommendation algorithm, which considers social tags as an important role forinformation retrieval. Besides its low cost of computational time, theexperiment results of two real-world data sets, \emph{Del.icio.us} and\emph{MovieLens}, show it can enhance the algorithmic accuracy and diversity.Especially, it can obtain more personalized recommendation results when usershave diverse topics of tags. In addition, the numerical results on thedependence of algorithmic accuracy indicates that the proposed algorithm isparticularly effective for small degree objects, which reminds us of thewell-known \emph{cold-start} problem in recommender systems. Further empiricalstudy shows that the proposed algorithm can significantly solve this problem insocial tagging systems with heterogeneous object degree distributions.



Author: Zi-Ke Zhang, Chuang Liu, Yi-Cheng Zhang, Tao Zhou

Source: https://arxiv.org/



DOWNLOAD PDF




Related documents