Learning Multiple Graphs for Document Recommendations
The Web offers rich relational data with di.erent semantics. In this paper, we address the problem of document recommendation in a digital library, where the documents in question are networked by citations and are associated with other entities by various relations. Due to the sparsity of a single graph and noise in graph construction, we propose a new method for combining multiple graphs to measure document similarities, where di.erent factorization strategies are used based on the nature of different graphs. In particular, the new method seeks a single low-dimensional embedding of documents that captures their relative similarities in a latent space. Based on the obtained embedding, a new recommendation framework is developed using semisupervised learning on graphs. In addition, we address the scalability issue and propose an incremental algorithm. The new incremental method signi.cantly improves the efficiency by calculating the embedding for new incoming documents only. The new batch and incremental methods are evaluated on two real world datasets prepared from CiteSeer. Experiments demonstrate signi.cant quality improvement for our batch method and signi.cant efficiency improvement with tolerable quality loss for our incremental method.
Recommender Systems Collaborative Filtering Semi-supervised Learning Social Network Analysis Spectral Clustering
Ding Zhou Shenghuo Zhu Xiaodan Song Belle L. Tseng Hongyuan Zha C. Lee Giles Kai Yu
Facebook Inc.156 University Avenue Palo Alto, CA 94301 NEC Labs America 10080 N Wolfe Road,Cupertino, CA 95014 Google Inc.1600 Amphitheatre Pkway,Mountain View, CA 94043 Yahoo! Inc.701 First Avenue Sunnyvale, CA 94089 College of Computing Georgia Institute of Technology Atlanta, GA 30332 Information Sciences andTechnology Computer Science & Engineering Pennsylvania State University Univ
国际会议
第十七届国际万维网大会(the 17th International World Wide Web Conference)(WWW08)
北京
英文
2008-04-21（万方平台首次上网日期，不代表论文的发表时间）