Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지
DOI QR코드

DOI QR Code

Input Variable Selection by Using Fixed-Point ICA and Adaptive Partition Mutual Information Estimation

고정점 알고리즘의 독립성분분석과 적응분할의 상호정보 추정에 의한 입력변수선택

  • 조용현 (대구가톨릭대학교 컴퓨터정보통신공학부)
  • Published : 2006.10.25
Download PDF
⟨ Previous Next ⟩

Abstract

This paper presents an efficient input variable selection method using both fixed-point independent component analysis(FP-ICA) and adaptive partition mutual information(AP-MI) estimation. FP-ICA which is based on secant method, is applied to quickly find the independence between input variables. AP-MI estimation is also applied to estimate an accurate dependence information by equally partitioning the samples of input variable for calculating the probability density function(PDF). The proposed method has been applied to 2 problems for selecting the input variables, which are the 7 artificial signals of 500 samples and the 24 environmental pollution signals of 55 samples, respectively The experimental results show that the proposed methods has a fast and accurate selection performance. The proposed method has also respectively better performance than AP-MI estimation without the FP-ICA and regular partition MI estimation.

본 논문에서는 고정점 알고리즘의 독립성분분석과 적응분할의 상호정보 추정을 조합한 입력변수선택 기법을 제안하였다. 여기서 고정점 알고리즘의 독립성분분석은 할선법에 기반을 둔 방법으로 입력변수 간의 독립성을 빠르게 찾기 위함이고, 적응분할의 상호정보 추정은 입력변수의 확률밀도함수 계산에서 동일한 량의 샘플분할을 가능하게 하여 변수상호간의 종속성을 좀 더 정확하게 구하기 위함이다. 제안된 기법을 인위적으로 제시된 각 500개의 샘플을 가지는 7개의 신호와 특정지역을 대상으로 측정된 각 55개의 샘플을 가진 24개의 환경오염신호를 대상으로 실험한 결과, 빠르고 정확한 변수의 선택이 이루어짐을 확인하였다. 또한 할선법의 고정점 알고리즘 독립성분분석을 수행하지 않을 때와 정규분할의 상호정보 추정 때보다 각각 우수한 선택성능이 있음을 확인하였다.

Keywords

References

  1. T. Trappenberg, J. Ouyang, and A. Back, 'Input Variable Selection : Mutual Information and Linear Mixing Measures,' IEEE Transactions on Knowledge and Data Engineering, Vol.1, No.8, pp. 37-46, Jan https://doi.org/10.1109/TKDE.2006.11
  2. A. Back and T. Trappenberg, 'Input Variable Selection Using Independent Component Analysis,' IJCNN99, pp. 1-5, Washington, 1999
  3. B. Blinnikiov and A. Weigend, 'Selecting Input Variables Using Mutual Information and Nonparametric Density Estimation,' Pro. of ISANN'94, pp. 42-50, Taiwan, Oct. 1994
  4. A. Back and A. Cichocki, 'Input Variable Selection Using Independent Component Analysis and Higher Order Statistics', Proc. of ICA99, Jan. 1999
  5. A. Back and T. Trappenberg, 'Selecting Inputs for Modelling Using Normalized Higher Order Statistics and Independent Component Analysis,' IEEE Transactions on Neural Networks, Vol.12, No.3, pp. 612-617, March. 2001 https://doi.org/10.1109/72.925564
  6. K. Atkinson, Elementary Numerical Analysis, John Wiley & Sons, Inc., New York, 1993
  7. A. Hyvarinen, J. Karhunen, and E. Oja, Independent Component Analysis, John Wiley & Sons, Inc., New York, May 2001
  8. T. W. Lee, Independent Component Analysis : Theory and Applications, Kluwer Academic Pub., Boston, Dec. 1998
  9. A. Hyvarinen and E. Oja, 'A Fast Fixed Point Algorithms for Independent Component Analysis,' Neural Computation, 9(7), pp. 1483-1492, Oct. 1997 https://doi.org/10.1162/neco.1997.9.7.1483
  10. J. Karhunen, 'Neural Approaches to Independent Component Analysis and Source Separation,' ESANN96, Burges, Belgium, pp. 249-266, Apr. 1996
  11. J. Wesley Hines, MATLAB Supplement to Fuzzy and Neural Approaches in Engineering, John Wiley & Sons, Inc., June 1997