chine model, called RBM-Cs, which can jointly model topic distribution of papers and 6 Conclusion In this paper, we formally define the problems of topic-based citation recommendation and propose a discriminative approach to this problem. Specifically, we proposes a two- layer Restricted Boltzmann Machine model, called RBM-CS, to model paper contents RBM-CS can significantly improve the recommendation performance There are many potential future directions of this work. It would be interesting to include other information for citation recommendation such as conference and author information. We are going to integrate the citation recommendation as a new feature intoouracademicsearchsystemArnetminer[10](http:/arnetminer.org) References 1. C. Buckley and E M. Voorhees. Retrieval evaluation with incomplete information. In Pro- ceedings of the 27th Annual International ACM SIGIR Conference on Research and Devel opment in Information Retrieval(SIGIR'04. pages 25-32, 2004 2. N Craswell, A P de Vries, and I Soboroff. Overview of the trec-2005 enterprise track. In TREC 2005 Conference Notebook, pages 199-205, 2005 3. E. Garfield. Citation analysis as a tool in journal evaluation. Science, 178(4060): 471-479 1972. 4. G. E. Hinton. Training products of experts by minimizing contrastive divergence. Neurai Computation,l4:1771-1800,2002. 5. G E Hinton. A fast learning algorithm for deep belief nets. Neural Computation, 18: 152 1554,2006 6. M. M. Kessler. Bibliographic coupling between scientific papers. American Documentation, 14:10-25.1963 7. S M. McNee, I. Albert, D Cosley, P Gopalkrishnan, S.K. Lam, A M. Rashid, J. A. Konstan, and J Riedl. On the recommending of citations for research papers. In CSCw02, page 116-125.2002. 8. P. Smolensky. Information processing in dynamical systems: foundations of harmony theory pages 194281,1986. In Proceedings of the 30th Annual Intemational ACM SIGIR Conference on Rese- papers 9. T Strohman, W. B. Croft, and D. Jensen. Recommending citations for academic papers Development in Information Retrieval (SIGIRO7, pages 705-706, 2007 10. J. Tang, J. Zhang, L. Yao, J. Li, L. Zhang, and z. Su. Arnetminer: Extraction and mining of academic social networks. In KDD'08, pages 990-998. 2008. 11. M. Welling. M. Rosen-zvi, and G. E Hinton. Exponential family harmoniums with an ap- plication to infomration retrieval. In Proceedings of the 17th Neural information Processing Systems(NIPS05), 2005 12. E. P. Xing, R. Yan, and A. G. Hauptmann. Mining associated text and images with dual-wing harmoniums. In Proceedings of the 21st Conference on Uncertainty in Artificial Intelligence UA'O5, pages633-641,20058 Jie Tang and Jing Zhang chine model, called RBM-CS, which can jointly model topic distribution of papers and citation relationships. 6 Conclusion In this paper, we formally define the problems of topic-based citation recommendation and propose a discriminative approach to this problem. Specifically, we proposes a two￾layer Restricted Boltzmann Machine model, called RBM-CS, to model paper contents and citation relationships simultaneously. Experimental results show that the proposed RBM-CS can significantly improve the recommendation performance. There are many potential future directions of this work. It would be interesting to include other information for citation recommendation, such as conference and author information. We are going to integrate the citation recommendation as a new feature into our academic search system ArnetMiner [10] (http://arnetminer.org). References 1. C. Buckley and E. M. Voorhees. Retrieval evaluation with incomplete information. In Pro￾ceedings of the 27th Annual International ACM SIGIR Conference on Research and Devel￾opment in Information Retrieval (SIGIR’04), pages 25–32, 2004. 2. N. Craswell, A. P. de Vries, and I. Soboroff. Overview of the trec-2005 enterprise track. In TREC 2005 Conference Notebook, pages 199–205, 2005. 3. E. Garfield. Citation analysis as a tool in journal evaluation. Science, 178(4060):471–479, 1972. 4. G. E. Hinton. Training products of experts by minimizing contrastive divergence. Neural Computation, 14:1771–1800, 2002. 5. G. E. Hinton. A fast learning algorithm for deep belief nets. Neural Computation, 18:1527– 1554, 2006. 6. M. M. Kessler. Bibliographic coupling between scientific papers. American Documentation, 14:10–25, 1963. 7. S. M. McNee, I. Albert, D. Cosley, P. Gopalkrishnan, S. K. Lam, A. M. Rashid, J. A. Konstan, and J. Riedl. On the recommending of citations for research papers. In CSCW’02, pages 116–125, 2002. 8. P. Smolensky. Information processing in dynamical systems: foundations of harmony theory. pages 194–281, 1986. 9. T. Strohman, W. B. Croft, and D. Jensen. Recommending citations for academic papers. In Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR’07), pages 705–706, 2007. 10. J. Tang, J. Zhang, L. Yao, J. Li, L. Zhang, and Z. Su. Arnetminer: Extraction and mining of academic social networks. In KDD’08, pages 990–998, 2008. 11. M. Welling, M. Rosen-Zvi, and G. E. Hinton. Exponential family harmoniums with an ap￾plication to infomration retrieval. In Proceedings of the 17th Neural Information Processing Systems (NIPS’05), 2005. 12. E. P. Xing, R. Yan, and A. G. Hauptmann. Mining associated text and images with dual-wing harmoniums. In Proceedings of the 21st Conference on Uncertainty in Artificial Intelligence (UAI’05), pages 633–641, 2005