Earticle

초록

영어

The paper is concerned with an experience - consistent development of fuzzy rule - based systems. This design of such fuzzy models involves some locally available data and then reconciles the constructed model with some previously acquired domain knowledge. This type of domain knowledge is captured in the format of several rule-based models constructed on a basis of some auxiliary data sets. To emphasize the nature of modeling being guided by this reconciliation mechanism, we refer to the resulting fuzzy model as experience – consistent identification. By forming a certain extended form of the optimized performance index, it is shown that the domain knowledge captured by the individual rule-based models play a similar role as a regularization component typically encountered in identification problems. We will show that a level of achieved experience-driven consistency can be quantified through fuzzy sets (fuzzy numbers) of the parameters of the local models standing in the conclusion parts of the rules. Experimental results involve both synthetic low - dimensional data and selected data coming from data available on the Web.

목차

Abstract
 1. Introduction and problem statement
 2. The experience-consistent development of the rule-based model
  2.1. The construction of information granules of conditions of the rules
  2.2. The consistency-based optimization of local regression models
  2.3. The alignment of information granules
 3. Characterization of experience-consistent models through its granular parameters
 4. Experimental studies
 5. Conclusions
 References

키워드

저자

• Partab Rai [ Department of Electrical & Computer Engineering University of Alberta ]
• Witold Pedrycz [ System Research Institute, Polish Academy of Sciences Warsaw, Poland ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Grid and Distributed Computing

• 간기

  격월간

• pISSN

  2005-4262

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605