Journal of the Institute of Electronics Engineers of Korea CI (전자공학회논문지CI)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

The Modified LVQ method for Performance Improvement of Pattern Classification

패턴 분류 성능을 개선하기 위한 수정된 LVQ 방식

  • Eom Ki-Hwan (Department of Electronic Engineering, Dongguk University) ;
  • Jung Kyung-Kwon (Department of Electronic Engineering, Dongguk University) ;
  • Chung Sung-Boo (Department of Applied Computer Added System, Seoil College)
  • Published : 2006.03.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper presents the modified LVQ method for performance improvement of pattern classification. The proposed method uses the skewness of probability distribution between the input vectors and the reference vectors. During training, the reference vectors are closest to the input vectors using the probabilistic distribution of the input vectors, and they are positioned to approximate the decision surfaces of the theoretical Bayes classifier. In order to verify the effectiveness of the proposed method, we performed experiments on the Gaussian distribution data set, and the Fisher's IRIS data set. The experimental results show that the proposed method considerably improves on the performance of the LVQ1, LVQ2, and GLVQ.

본 논문에서는 수정된 LVQ를 이용한 패턴 분류 방식을 제안한다. 제안한 방식은 입력 패턴의 분류 성능을 개선하기 위하여 입력 벡터와 기준 벡터 사이의 확률 분포의 비대칭도를 계산하여 학습에 이용한다. 학습을 하는 동안 기준 벡터는 입력 벡터의 확률 분포에 근접하게 되고, 기준 벡터는 Bayes 분류기의 결정 경계에 근접하게 위치한다. 가우시안 분포의 데이터와 Fisher의 IRIS 데이터 분류를 실험하여 LVQ1, LVQ2, GLVQ와 비교하여 제안한 방식이 우수한 분류 성능을 나타냄을 확인하였다.

Keywords

References

  1. Carl G. Looney, Pattern Recognition Using Neural Network, Oxford Univ. Press, 1997
  2. Martin T. Hagan, and Howard B. Demuth, Mark H. Beale, Neural Network Design, Thomson Learning, 1995
  3. T. Kohonen, 'The self organizing map,' Proceedings of the IEEE, vol. 78, no. 9, pp. 1464-1480, 1990 https://doi.org/10.1109/5.58325
  4. T. Kohonen, 'The self organizing map,' Neurocomputing, vol. 21, no. 1 3, pp. 1 6, 1998 https://doi.org/10.1016/S0925-2312(98)00030-7
  5. Simon Haykin, Neural Networks, 2/E: A Comprehensive Foundation, Prentice Hall, 1998
  6. A. I. Gonzalez, M. Grana, and A. D'Anjou, 'An analysis of the GLVQ algorithm,' IEEE Transactions on Neural Networks, vol. 6, no. 4, pp.l012 1016, 1995 https://doi.org/10.1109/72.392266
  7. S. C. Ahalt, A. K. Krishnamurthy, P. Chen, and D. E. Melton, 'Competitive Learning Algorithm for Vector Quantization,' Neural Networks, vol. 3, pp. 277 290, 1990 https://doi.org/10.1016/0893-6080(90)90071-R
  8. Allen Gersho, and Robert M. Gray, Vector Quantization and Signal Compression, Kluwer Academic Pub., 1992
  9. S. P. Luttrell, 'Code vector density in topographic mappings: Scalar case,' IEEE Transactions on Neural Networks, vol. 2, no. 4, pp. 427 436, 1991 https://doi.org/10.1109/72.88162
  10. H. Ritter, 'Asymptotic level density for a class of vector quantization processes,' IEEE Transactions on Neural Networks, vol. 2, no. 1, pp. 173 175, 1991 https://doi.org/10.1109/72.80310
  11. N. Kitajima, 'A new method for initializing reference vectors in LVQ,' IEEE Int. Conf. on Neural Networks, vol. 5, pp. 2775 -2779, Nov/Dec, 1995 https://doi.org/10.1109/ICNN.1995.488170
  12. 정경권, 이용구, 엄기환, 'SOM을 이용한 LVQ 네트워크 설계', 전자공학회논문지 제40권 CI편, 제5호, 280-288쪽, 2003년 9월
  13. Michio Kaku, Hyperspace, Random House, 1995
  14. D. R. Smith, Variational Methods in Optimization, Dover, 1998