Journal of KIISE:Software and Applications (한국정보과학회논문지:소프트웨어및응용)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

Combining Multiple Classifiers using Product Approximation based on Third-order Dependency

3차 의존관계에 기반한 곱 근사를 이용한 다수 인식기의 결합

  • Published : 2004.05.01
Download PDF
⟨ Previous Next ⟩

Abstract

Storing and estimating the high order probability distribution of classifiers and class labels is exponentially complex and unmanageable without an assumption or an approximation, so we rely on an approximation scheme using the dependency. In this paper, as an extended study of the second-order dependency-based approximation, the probability distribution is optimally approximated by the third-order dependency. The proposed third-order dependency-based approximation is applied to the combination of multiple classifiers recognizing handwritten numerals from Concordia University and the University of California, Irvine and its usefulness is demonstrated through the experiments.

인식기와 클래스 레이블로 구성된 고차 확률 분포를 가정이나 근사 없이 저장하고, 평가하는 것은 기하급수적으로 복잡하고 관리하기 어렵다. 따라서, 의존관계를 이용한 근사 방법에 의존하게 된다. 본 논문에서는 기존의 2차 의존관계에 기반 한 곱 근사 방법을 확장하여, 이 확률 분포를 3차 의존관계에 의해 최적으로 곱 근사 하는 방법을 제안하고자 한다. 제안된 3차 의존관계에 기반 한 곱 근사 방법은 Concordia 대학과 UCI(University of California, Irvine) 대학으로부터 얻은 필기 숫자를 인식하는 실험에서 다수 인식기의 결합 방법에 적용되었고, 실험을 통하여 제안된 방법의 유용성을 살펴보았다.

Keywords

References

  1. L. Xu and A. Krzyzak, and C. Y. Suen, 'Methods of Combining Multiple Classifiers and Their Applications to Handwriting Recognition,' IEEE Trans. on Systems, Man and Cybernetics, vol.22, no.3, pp.418-435, 1992 https://doi.org/10.1109/21.155943
  2. 강희중, 이성환, '무제약 필기 숫자를 인식하기 위한 다수 인식기를 결합하는 의존관계 기반의 프레임워크', 정보과학회논문지 : 소프트웨어 및 응용, 제27건, 제8호, pp.855-863, 2000
  3. Y. S. Huang and C. Y. Suen, 'A Method of Combining Multiple Experts for the Recognition of Unconstrained Handwritten Numerals,' IEEE Trans. on Pattern Analysis and Machine Intelligence, vol.17, no.1, pp.90-94, 1995 https://doi.org/10.1109/34.368145
  4. Mandler, E. and Schuermann, J., 'Combining the classification results of independent classifiers based on the dempster/shafer theory of evidence,' In E. S. Gelsema and L. N. Kanal, editors, Pattern Recognition and Artifical Intelligence, pp. 381-393, 1988
  5. D.-S. Lee and S. N. Srihari, 'Handprinted Digit Recognition : A Comparison of Algorithms,' Proc. of the 3rd Int. Workshop on Frontiers in Handwriting Recognition, pp.42-45, 1993
  6. C. K. Chow and C. N. Liu, 'Approximating Discrete Probability Distributions with Dependence Trees,' IEEE Trans. on Information Theory, vol.14, no.3, pp.462-467, 1968 https://doi.org/10.1109/TIT.1968.1054142
  7. P. M. Lewis, 'Approximating Probability Distributions to Reduce Strange Requirement,' Information and Control, vol.2, pp.214-225, Sep., 1959 https://doi.org/10.1016/S0019-9958(59)90207-4
  8. Matsui, T., Takamura, T., Srihari, S. N., 'Combination of stroke/Background Structure and Contour-direction Features in Handprinted Alphanumeric Recognition,' In Proceedings of the 4th Int. Workshop on Frontiers in Handwriting Recognition, pp. 87-96, 1994
  9. Oh, I.-S. and Suen, C. Y., 'Distance features for neural network-based recognition of handwritten characters,' International Journal on Document Analysis and Recognition, 1(2):73-80, 1998 https://doi.org/10.1007/s100320050008
  10. Oh, I.-S., Lee, I.-S., Hong, K.-S. and Choi, S.-M., 'Class-expert approach use unconstrained handwritten numeral recognition,' Proceedings of the 5th Int. Workshop on Frontiers in Handwriting Recognition, pp.35-40, 1996
  11. Suen, C. Y., Nidal, C., Legault, R., Mai, T. A., and Lam, L., 'Computer Recognition of Unconstrained Handwritten Numerals,' In Proceedings of IEEE, pp. 1162-1180, 1992 https://doi.org/10.1109/5.156477
  12. Blake, C. and Merz, C., UCI repository of machine learning databases [http://www.ics.uci.edu/~mlearn/mlrepository.html], Irvine, CA, Dept. of Information and Computer Sciences, 1998