|Table of Contents|

[1] Xing Zongyi, Hu Weili, Jia Limin,. Study on the tradeoff between interpretabilityand precision in fuzzy modeling [J]. Journal of Southeast University (English Edition), 2004, 20 (4): 472-476. [doi:10.3969/j.issn.1003-7985.2004.04.016]
Copy

Study on the tradeoff between interpretabilityand precision in fuzzy modeling()
一种模糊建模方法的研究: 精确性与解释性的折衷
Share：

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:
20
Issue:
2004 4
Page:
472-476
Research Field:
Automation
Publishing date:
2004-12-30

Info

Title:
Study on the tradeoff between interpretabilityand precision in fuzzy modeling
一种模糊建模方法的研究: 精确性与解释性的折衷
Author(s):
Xing Zongyi1;  Hu Weili1;  Jia Limin2
1Department of Automation, Nanjing University of Science and Technology, Nanjing 210094, China
2School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China
邢宗义1;  胡维礼1;  贾利民2
1南京理工大学自动化系, 南京 210094; 2北京交通大学交通运输学院, 北京 100044
Keywords:
fuzzy modeling;  precision;  interpretability;  fuzzy clustering
模糊建模;  精确性;  解释性;  模糊聚类
PACS:
TP273
DOI:
10.3969/j.issn.1003-7985.2004.04.016
Abstract:
An approach to identifying fuzzy models considering both interpretability and precision is proposed. Firstly, interpretability issues about fuzzy models are analyzed. Then, a heuristic strategy is used to select input variables by increasing the number of input variables, and the Gustafson-Kessel fuzzy clustering algorithm, combined with the least square method, is used to identify the fuzzy model. Subsequently, an interpretability measure is described by the product of the number of input variables and the number of rules, while precision is weighted by root mean square error, and the selection objective function concerning interpretability and precision is defined. Given the maximum and minimum number of input variables and rules, a set of fuzzy models is constructed. Finally, the optimal fuzzy model is selected by the objective function, and is optimized by a genetic algorithm to achieve a good tradeoff between interpretability and precision. The performance of the proposed method is illustrated by the well-known Box-Jenkins gas furnace benchmark; the results demonstrate its validity.
提出一种同时考虑解释性和精确性的模糊建模方法. 首先分析影响模糊模型解释性的主要因素, 然后利用启发式搜索策略实现输入变量选择, 利用模糊聚类算法和最小二乘辨识模糊模型. 随后以输入变量数目和模糊规则数目的乘积衡量可解释性, 以均方误差衡量精确性, 并据此定义模型选择目标函数. 最后给定最大最小的输入变量数目和规则数目, 辨识得到一组模糊模型, 利用模型选择目标函数, 选择最优的模糊模型, 并采用遗传算法进行优化, 达到解释性与精确性的折衷. 煤气炉仿真例子验证了该方法的有效性.

References:

[1] Gomez-Skarmeta A F, Delgado M, Vila M A. About the use of fuzzy clustering techniques for fuzzy model identification [J]. Fuzzy Sets and Systems, 1999, 106(2): 179-188.
[2] Tsoukalas L H, Uhrig R E. Fuzzy and neural approaches in engineering [M]. New York: Wiley, 1997.
[3] Cordon O, Herrera F, Hoffmann F, et al. Genetic fuzzy systems: evolutionary tuning and learning of fuzzy rule bases [M]. Singapore: World Scientific, 2001.
[4] Babuska R, Bersini H, Linkens D A, et al. Future prospects for fuzzy systems and technology [EB/OL].http: //www.erudit.de/erudit/newsletters/news61/pag e5.htm. 2000-11-21/2004-09-17.
[5] Takagi T, Sugeno M. Fuzzy identification of systems and its application to modeling and control [J]. IEEE Trans on Systems, Man and Cybernetics, 1985, 15(1): 116-132.
[6] Xing Zongyi, Hu Weili, Jia Limin. A fuzzy clustering based approach for generating interpretable fuzzy models [A]. In: Proc of the Third International Conference on Machine Learning and Cybernetics [C]. Shanghai, 2004. 2093-2097.
[7] Xing Zongyi, Jia Limin. Research on input variable selection for numeric data based fuzzy modeling [A]. In: Proc of Int Conf on Machine Learning and Cybernetics [C]. Xi’an, 2003. 2737-2740.
[8] Bezdek J C. Pattern recognition with fuzzy objective algorithm [M]. New York: Plenum Press, 1981.
[9] Gustafson D E, Kessel W C. Fuzzy clustering with a fuzzy covariance matrix [A]. In: Proc of IEEE Conf on Decision and Control [C]. San Diego, 1979. 761-766.
[10] Setnes M, Roubos H. Transparent fuzzy modeling using fuzzy clustering and GA’s. [A]. In: Proc of IEEE Int Conf on Fuzzy Systems [C]. Seou, 1999. 198-202.

Memo

Memo:
Biographies: Xing Zongyi(1974—), male, doctor, xingzongyi@com.cn; Hu Weili(1941—), male, professor, hwl1002@mail.njust.edu.cn.
Last Update: 2004-12-20