Combining Convolution Kernels Defined on Heterogeneous Sub-structures

  • Minlie Huang
  • Xiaoyan Zhu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4426)


Convolution kernels, constructed by convolution of sub-kernels defined on sub-structures of composite objects, are widely used in classification, where one important issue is to choose adequate sub-structures, particularly for objects such as trees, graphs, and sequences. In this paper, we study the problem of sub-structure selection for constructing convolution kernels by combining heterogeneous kernels defined on different levels of sub-structures. Sub-kernels defined on different levels of sub-structures are combined together to incorporate their individual strengths because each level of sub-structure reflects its own angle to view the object. Two types of combination, linear and polynomial combination, are investigated. We analyze from the perspective of feature space why combined kernels exhibit potential advantages. Experiments indicate that the method will be helpful for combining kernels defined on arbitrary levels of sub-structures.


SVM convolution kernel text mining relation extraction 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Scholkopf, B.: Support vector learning. Oldenbourg, München (1997)Google Scholar
  2. 2.
    Haussler, D.: Convolution kernels on discrete structures. Technical report, UC Santa CruzGoogle Scholar
  3. 3.
    Collins, M.: New Ranking Algorithms for Parsing and Tagging: Kernels over Discrete Structures, and the Voted Perceptron. In: ACL 2002, pp. 263–270 (2002)Google Scholar
  4. 4.
    Zelenko, D., Aone, C., Richardella, A.: Kernel Methods for Relation Extraction. Journal of Machine Learning Research 3, 1083–1106 (2003)zbMATHCrossRefMathSciNetGoogle Scholar
  5. 5.
    Lodhi, H., et al.: Text Classification Using String Kernels. Journal of Machine Learning Research 2(2), 419–444 (2002)zbMATHCrossRefGoogle Scholar
  6. 6.
    Bunescu, R., Mooney, R.: A Shortest Path Dependency Kernel for Relation Extraction. In: EMNLP, Vancouver, B.C., pp. 724-731Google Scholar
  7. 7.
    Joachims, T., Cristianini, N., Shawe-Taylor, J.: Composite Kernels for Hypertext Categorisation. In: Proceedings of ICML-01, 18th ICML, pp. 250–257 (2001)Google Scholar
  8. 8.
    Lanckriet, G., et al.: Kernel-based Data Fusion and its Application to Protein Function Prediction in Yeast. In: PSB, pp. 300–311 (2004)Google Scholar
  9. 9.
    Zhao, S., Grishman, R.: Extracting Relations with Integrated Information Using Kernel Methods. In: ACL 2005, Ann Arbor, June 2005, pp. 419–426 (2005)Google Scholar
  10. 10.
    Lin, D.: A Dependency-based Method for Evaluating Broad-Coverage Parsers. In: ICAI 1995, pp. 1420–1427 (1995)Google Scholar
  11. 11.
    Ohta, Y., et al.: The GENIA Corpus: An Annotated Research Abstract Corpus in the Molecular Biology Domain. In: Human Language Technologies Conference, pp. 73–77 (2002)Google Scholar
  12. 12.
    Joachims, T.: Text Categorization with Support Vector Machines: Learning with Many Relevant Features. In: Nédellec, C., Rouveirol, C. (eds.) ECML 1998. LNCS, vol. 1398, pp. 137–142. Springer, Heidelberg (1998)CrossRefGoogle Scholar

Copyright information

© Springer Berlin Heidelberg 2007

Authors and Affiliations

  • Minlie Huang
    • 1
  • Xiaoyan Zhu
    • 1
  1. 1.Department of Computer Science and Technology, Tsinghua University, Beijing,100084China

Personalised recommendations