摘要: 提出一种新的参数优化模型和求解算法,引入模糊熵来指导模糊产生式规则(FPRs)的参数优化,给出基于极大模糊熵定理的参数优化模型,提出求解该模型的Pareto强度值的演化算法。实验结果表明,该方法能够有效优化参数,一定程度上避免过度拟合,提高了FPRs的知识表示能力。
关键词:
模糊产生式规则,
知识表示参数,
极大模糊熵定理,
Pareto强度值演化
Abstract: This paper introduces fuzzy entropy into the procedure of exploring parameters of Fuzzy Production Rules(FPRs). A parameter optimization model based on maximum fuzzy entropy principle is proposed and a Pareto strength evolutionary algorithm is introduced to solve this model. Experimental results show that the trained parameters gained from above strategy are highly accurate, therefore this method can decrease the phenomenon of over-fitting and improve the knowledge representation capability of FPRs.
Key words:
Fuzzy Production Rules(FPRs),
knowledge representation parameter,
maximum fuzzy entropy principle,
Pareto strength evolution
中图分类号:
安素芳;柴变芳;傅 玥;才秀凤. 基于Pareto强度值演化的FPRs知识表示参数优化[J]. 计算机工程, 2008, 34(8): 218-220.
AN Su-fang; CHAI Bian-fang; FU Yue; CAI Xiu-feng. Knowledge Representation Parameters Optimization of FPRs Based on Pareto Strength Evolution[J]. Computer Engineering, 2008, 34(8): 218-220.