Journal of Intelligent Information Systems

, Volume 26, Issue 3, pp 269–301 | Cite as

Latent linkage semantic kernels for collective classification of link data

Article

Abstract

Generally, links among objects demonstrate certain patterns and contain rich semantic clues. These important clues can be used to improve classification accuracy. However, many real-world link data may exhibit more complex regularity. For example, there may be some noisy links that carry no human editorial endorsement about semantic relationships. To effectively capture such regularity, this paper proposes latent linkage semantic kernels (LLSKs) by first introducing the linkage kernels to model the local and global dependency structure of a link graph and then applying the singular value decomposition (SVD) in the kernel-induced space. For the computational efficiency on large datasets, we also develop a block-based algorithm for LLSKs. A kernel-based contextual dependency network (KCDN) model is then presented to exploit the dependencies in a network of objects for collective classification. We provide experimental results demonstrating that the KCDN model, together with LLSKs, demonstrates relatively high robustness on the datasets with the complex link regularity, and the block-based computation method can scale well with varying sizes of the problem.

Keywords

Kernel methods Link regularity Latent linkage semantic kernel Kernel-based contextual dependency networks Block-based link analysis Collective classification 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Bernstein, A., Clearwater, S., & Provost, F. (2003). The relational vector-space model and industry classification. In Proc. IJCAI-2003 Workshop on Learning Statistical Models from Relational Data (pp. 8–18).Google Scholar
  2. Chakrabarti, S., Dom, B., & Indyk, P. (1998). Enhanced hypertext categorization using hyperlinks. SIGMOD Record, 27(2), 307–318.CrossRefGoogle Scholar
  3. Chen, Z., Liu, S. P., Liu, W. Y., Pu, G. G., & Ma,W. Y. (2003). Building a web thesaurus from web link structure. In Proc. 26th Annual Int’l ACM SIGIR Conf. Research and Development in Information Retrieval(SIGIR’03) (pp. 48–55). Toronto, Canada: ACM Press, New York, NY, USA.Google Scholar
  4. Craven, M., DiPasquo, D., Freitag, D. McCallum, A., Mitechell, T., Nigam, K., & Slattery, S. (1998). Learning to extract symbolic knowledge from the world wide web. In Proc. 15th National Conf. Artificial Intelligence (AAAI-98) (pp. 509–516). Madison, US: AAAI Press, Menlo Park, US.Google Scholar
  5. Cristianini, N., Shawe-Talyor, J., & Lodhi, H. (2002). Latent semantic kernels. Journal of Intelligent Information Systems, 18(2/3), 127–152.CrossRefGoogle Scholar
  6. Datta, B. N. (1995). Numerical linear algebra and application. Brooks/Cole Publishing Co., Pacific Grove, CA.Google Scholar
  7. Davison, B. (2000). Recognizing nepotistic links on the web. In Proc. AAAI-2000 Workshop on Artificial Intelligence for Web Search (pp. 23–28). Austin, Texas: AAAI Press, Menlo Park, US.Google Scholar
  8. Deerwester, S., Dumais, S. T., Furnas, G. W., Landauer, T. K., & Harshman, R. (1990). Indexing by latent semantic analysis. Journal of the American Society for Information Science, 41(6), 389–401.CrossRefGoogle Scholar
  9. Drineas, P., Kannan, R., Frieze, A., & Vinay, V. (2004). Clustering large graphs via the Singular Value Decomposition. Machine Learning, 56, 9–33.MATHCrossRefGoogle Scholar
  10. Fisher, M. J., & Everson, R. M. (2003). When are links useful? Experiments in text classification. In F. Sebastini (Ed.), Advances in information retrieval. 25th European Conference on IR Research (ECIR 2003) (pp. 41–56). Pisa, Italy: Springer.Google Scholar
  11. Gärtner, T. (2003). A survey of kernels fro structured data. SIGKDD Explorations Newsletter, 5(1), 49–58.Google Scholar
  12. Getoor, L., Friedman, N., Koller, D., & Taskar, B. (2002). Learning probabilistic models of relational structure. Journal of Machine Learning Research, 1, 679–707.Google Scholar
  13. Hawking, D., Voorhees, E., Craswell, N., & Bailey, P. (1999). Overview of the TREC-8 web track. In E. M. Voorhees and D. K. Harman (Eds.), Proc. 8th Text REtrieval Conf. (TREC-8), (pp. 131–150).Google Scholar
  14. Heckerman, D., Chickering, D., Meek, C., Rounthwaite, R., & Kadie, C. (2001). Dependency networks for inference, collaborative filtering, and data visualization. Journal of Machine Learning Research, 1, 49–75.MATHCrossRefGoogle Scholar
  15. Henzinger, M. R. (2001, Jan–Feb). Hyperlink analysis for the web. IEEE Internet Computing, 5(1), 45–50.CrossRefGoogle Scholar
  16. Hou, J., & Zhang, Y. (2003). Effectively finding relevant Web pages from linkage information. IEEE Transaction on Knowledge and Data Engineering, 15(4), 940–951.CrossRefGoogle Scholar
  17. Jensen, D., & Neville, J. (2002). Linkage and autocorrelation cause feature selection bias in relational learning. In Proc. 9th Int’l Conf. Machine Learning (ICML03) (pp. 259–266). Morgan Kaufmann Publishers: San Francisco, USA.Google Scholar
  18. Joachims, T., Cristianini, N., & Shawe-Talyor, J. (2001). Composite kernels for hypertext categorization. In Proc. 8th Int’l Conf. Machine Learning (ICML01) (pp. 250–257). San Francisco, US: Morgan Kaunfmann Publishers.Google Scholar
  19. Kamvar, S. D., Haveliwala, T. H., Manning, C., & Golub, G. H. (2003). Exploiting the block structure of the Web for computing PageRank. Technical report, Department of Computer Science, Stanford University.Google Scholar
  20. Kandola, J., Shawe-Talyor, J., & Cristianini, N. (2002). Learning semantic similarity. In Int’l Conf. Advances in Information Processing System (NIPS 15). MIT Press: Cambridge, MA, USA.Google Scholar
  21. Kao, H.-Y., Lin, S.-H., Ho, J.-M., & Chen, M.-S. (2004). Mining Web informative structures and contents based on Entropy analysis. IEEE Transaction on Knowledge and Data Engineering, 18(1), 41–55.CrossRefGoogle Scholar
  22. Karypis, G., & Kumar, V. (1998). METIS, A Software Package for Partitioning Unstructured Graphs, Partitioning Meshes, and Computing Fill-Reducing Orderings of Sparse Matrices, Version 4.0, http://www.cs.umn.edu/~karypis/metis.
  23. Kleinberg, J. (1999). Authoritative sources in a hyperlinked environment. Journal of ACM, 46(5), 604–632.MATHMathSciNetCrossRefGoogle Scholar
  24. Kondor, R. I., & Lafferty, J. (2002). Diffusion kernels on graphs and other discrete structures. In C. Sammut & A. Hoffman (Eds), Proc. 19th Int’l Conf. Machine Learning (ICML01) (pp. 315–322). Morgan Kaufmann Publishers: San Francisco, USA.Google Scholar
  25. Lu, Q., & Getoor, L. (2003). Link-based classification. In Fawcett & N. Mishra (Eds.), Proc. 12th Int’l Conf. Machine Learning (ICML03) (pp. 496–503). Washington DC: AAAI Press, Menlo Park, US.Google Scholar
  26. McCallum, A., Nigam, K., Rennie, J., & Seymore, K. (2000). Automating the construction of internet portals with machine learning. Information Retrieval Journal, 3, 127–163.CrossRefGoogle Scholar
  27. Neville, J., & Jensen, D. (2003). Collective classification with relational dependency networks. In Proc. 2nd Multi-Relational Data Mining Workshop, 9th ACM SIGKDD Int’l Conf. Knowledge Discovery and Data Mining (pp. 77–91). Washington, DC, USA.Google Scholar
  28. Page, L., Brin, S., Motwani, R., & Winograd, T. (1998). The PageRank citation ranking: Bring order to the web. Technical report, Standford University.Google Scholar
  29. Pierre, J. M. (2001). On the automated classification of Web sites. Linköping Electronic Articles in Computer and Information Science, 6(001), Sweden.Google Scholar
  30. Richardson, M., & Domingos, P. (2004). Markov logic networks. Technical report, Department of Computer Science and Engineering, University of Washington, Seattle, WA. http://www.cs.washington.edu/homes/pedrod/mln.pdf.
  31. Schölkopf, B. (2000). Statistical learning and kernel methods. Technical Report, MSR-TR-2000-23, Microsoft Research.Google Scholar
  32. Schölkopf, B., Smola, A. J., & Müller, K.-R. (1998). Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation, 10, 1299–1319.CrossRefGoogle Scholar
  33. Simon, H. D., & Zha, H. (1999). Low-rank matrix approximation using the Lanczos bidiagonalization process with applications. SIAM Journal on Scientific Computing, 21(6), 2257–2274.MathSciNetCrossRefGoogle Scholar
  34. Taskar, B., Segal, E., & Koller, D. (2001). Probabilistic classification and clustering in relational data. In Proc. 17th Int’l Joint Conf. Artificial Intelligence(IJCAI01) (pp. 870–876). Seattle, Washington.Google Scholar
  35. Taskar, B., Abbeel, P., & Koller, D. (2002). Discriminative probabilistic models for relational classification. In Proc. Uncertainty on Artificial Intelligence (UAI-02) (pp. 485–492). Edmonton, Canada.Google Scholar
  36. Yang, Y., Slattery, S., & Ghani, R. (2002). A study of approaches to hypertext categorization. Journal of Intelligent Information System, 18(2/3), 219–241.CrossRefGoogle Scholar
  37. Zhong, S., & Ghosh, J. (2001). A new formulation of coupled hidden Markov models. Technical report, Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, United States.Google Scholar

Copyright information

© Springer Science + Business Media, LLC 2006

Authors and Affiliations

  1. 1.Institute of Computing TechnologyChinese Academy of SciencesBeijingPR China
  2. 2.Digital Media InstitutePeking UniversityBeijingPR China

Personalised recommendations