DOI QR코드

DOI QR Code

Selection Method of Fuzzy Partitions in Fuzzy Rule-Based Classification Systems

퍼지 규칙기반 분류시스템에서 퍼지 분할의 선택방법

  • 손창식 (영남대학교 전기공학과) ;
  • 정환묵 (대구가톨릭대학교 컴퓨터정보통신공학부) ;
  • 권순학 (영남대학교 전기공학과)
  • Published : 2008.06.25

Abstract

The initial fuzzy partitions in fuzzy rule-based classification systems are determined by considering the domain region of each attribute with the given data, and the optimal classification boundaries within the fuzzy partitions can be discovered by tuning their parameters using various learning processes such as neural network, genetic algorithm, and so on. In this paper, we propose a selection method for fuzzy partition based on statistical information to maximize the performance of pattern classification without learning processes where statistical information is used to extract the uncertainty regions (i.e., the regions which the classification boundaries in pattern classification problems are determined) in each input attribute from the numerical data. Moreover the methods for extracting the candidate rules which are associated with the partition intervals generated by statistical information and for minimizing the coupling problem between the candidate rules are additionally discussed. In order to show the effectiveness of the proposed method, we compared the classification accuracy of the proposed with those of conventional methods on the IRIS and New Thyroid Cancer data. From experimental results, we can confirm the fact that the proposed method only considering statistical information of the numerical patterns provides equal to or better classification accuracy than that of the conventional methods.

퍼지 규칙기반 분류 시스템에서 초기의 퍼지 분할은 주어진 데이터가 가진 속성들의 도메인을 고려함으로서 결정되어지고, 최적의 분류 경계면은 초기에 정의된 퍼지 분할의 파라미터들을 조정함으로서 찾을 수 있다. 본 논문에서는 학습과정들을 사용하지 않고 패턴분류의 성능을 최대화하기 위해 통계적 정보에 기반을 둔 퍼지 분할의 선택방법을 제안한다. 제안된 방법에서 통계적 정보는 주어진 수치적인 데이터로부터 각 입력 속성의 '불확실성 영역', 즉 패턴분류문제에서 분류 경계면이 결정되는 영역을 추출하기 위해 사용되었다. 또한 통계적인 정보에 의해서 생성된 퍼지 분할구간에 대응하는 후보 규칙들을 추출하기 위한 방법과 그 후보 규칙들 간의 커플링 문제를 최소화하기 위한 방법도 추가적으로 논의하였다. 실험에서는 제안된 방법의 효용성을 보이기 위해 IRIS와 New Thyroid Cancer 데이터를 사용한 기존 패턴분류 방법들과의 분류 정확성을 비교하였고, 그 결과들로부터 제안된 방법이 기존의 방법들보다 더 좋은 분류 정확성을 제공함을 확인할 수 있었다.

Keywords

References

  1. H. Ishibuchi, T. Nakashima, and T. Murata, "Three-objective genetics-based machine learning for linguistic rule extraction," Information sciences, vol. 136, no. 1-4, pp. 109-133, 2001 https://doi.org/10.1016/S0020-0255(01)00144-X
  2. H. Ishibuchi and T. Yamamoto, "Fuzzy rule selection by data mining criteria and genetic algorithms," Genetic evolutionary computation conf. (GECCO-2002), pp. 399-406, 2002
  3. H. Ishibuchi, K. Nozaki, N. Yamamoto, and H. Tanaka, "Construction of fuzzy classification systems with rectangular fuzzy rules using genetic algorithm," Fuzzy sets and systems, vol. 65, no. 2-3, pp. 237-253, 1994 https://doi.org/10.1016/0165-0114(94)90022-1
  4. J. Casillas, O.Cordon, M.J.Del Jesus, and F. Herrera, "Genetic feature selection in a fuzzy rule-based classification system learning process for high-dimensional problems," Information sciences, vol. 136, pp. 135-157, 2001 https://doi.org/10.1016/S0020-0255(01)00147-5
  5. V. Uebele, S. Abe, and M.S. Lan, "A neural network based fuzzy classifier," IEEE Transactions on SMC, vol. 25, pp. 353-361, 1995 https://doi.org/10.1109/21.364829
  6. S. Abe and M.S. Lan, "A method for fuzzy rules extraction directly from numerical data and its application to pattern classification," IEEE Transactions on fuzzy systems, vol. 3, pp. 18-28, 1995 https://doi.org/10.1109/91.366565
  7. M.A. Kbir, H. Benkirane, K. Maalmi, and R. Benslimane, "Hierarchical fuzzy partition for pattern classification with fuzzy if-then rules," Pattern recognition letters, vol. 21, no. 6-7, pp. 503-509, 2000 https://doi.org/10.1016/S0167-8655(00)00015-5
  8. 손창식, 정환묵, 서석태, 권순학, "규칙의 커플링 문제를 최소화하기 위한 퍼지-러프 분류 방법," 한국 퍼지 및 지능시스템 학회 논문지, vol. 17, no. 4, pp. 460-465, 2007 https://doi.org/10.5391/JKIIS.2007.17.4.460
  9. A. Skowron and C.M. Rauszer, "The Discernibility matrices and functions in information systems," Institute of computer sciences report 1/91, Technical University of Warsaw, pp. 1-41, 1991
  10. UCI Repository of Machine Learning Databases, Depatment of Information and Computer Science, University of California, Irvine, CA, Available: http://mlearn.ics.uci.edu/MLRepository.html
  11. D. Nauck, U. Nauck, and R. Kruse, "Generating classification rules with the neuro-fuzzy system NEFCLASS," In proceedings of the biennial conference of NAFIPS, Berkeley, pp. 19-22, 1996
  12. S.M. Nozaki, H. Ishibuchi, and H. Tanaka, "Adaptive fuzzy rule based classification systems," IEEE Transactions on fuzzy systems, vol. 4, no. 3, pp. 238-250, 1996 https://doi.org/10.1109/91.531768
  13. H. Ishibuchi, T. Murata, and I.B. Turksen, "Single objective and two-objective genetic algorithms for selecting linguistic rules for pattern classification problems," Fuzzy sets and systems, vol. 89, no. 2, pp. 135-150, 1997 https://doi.org/10.1016/S0165-0114(96)00098-X
  14. Y.C. Tsai, C.H. Cheng, and J.R. Chang, "Entropy-based fuzzy rough classification approach for extracting classification rules," Experts systems with applications, vol. 31, no. 2, pp. 436-443, 2006 https://doi.org/10.1016/j.eswa.2005.09.038
  15. N. R. Guo, T.H.S. Li, and C.L. Kuo, "Hierarchical fuzzy model for classification problem," IECON 2002, vol. 3, pp. 2096-2101, 2002
  16. Y. Chen, B. Yang, A. Abraham, and L. Peng, "Automatic design of hierarchical takagi-sugeno type fuzzy systems using evolutionary algorithms," IEEE transactions on fuzzy systems, vol. 15, no. 3, pp. 385-397, 2007 https://doi.org/10.1109/TFUZZ.2006.882472
  17. J-S. R. Jang, "ANFIS: Adaptive network based fuzzy inference systems," IEEE Transactions on SMC, vol. 23, no. 3, pp. 665-695, 1993 https://doi.org/10.1109/21.256541
  18. J.W. Chinneck, "Fast heuristics for the maximum feasible subsystem problem," INFORMS Journal on computing, vol. 13, no. 3, pp. 210-223, 2001 https://doi.org/10.1287/ijoc.13.3.210.12632
  19. R. Thawonmas and S. Abe, "A novel approach to feature selection based on analysis of class regions," IEEE Transactions on SMC, vol. 27, pp. 196-207, 1997