Title: Evolutionarily adjusting membership functions in Takagi-Sugeno fuzzy systems

Authors: Tzung-Pei Hong, Wei-Tee Lin, Chun-Hao Chen, Chen-Sen Ouyang

Addresses: Department of Computer Science and Information Engineering, National University of Kaohsiung, Kaohsiung, No. 700, Kaohsiung University Rd., Nanzih District, 811 Kaohsiung, Taiwan; Department of Computer Science and Engineering, National Sun Yat-sen University, Kaohsiung, No. 70, Lienhai Rd., Kaohsiung 80424, Taiwan. ' Department of Computer Science and Information Engineering, National Cheng-Kung University, No. 1, University Road, Tainan City 701, Taiwan. ' Department of Computer Science and Information Engineering, Tamkang University Taipei, No. 151, Ying-Chuan Road, Tamsui, Taipei County 25137, Taiwan. ' Department of Computer Science and Information Engineering, I-Shou University, Kaohsiung, No. 1, Sec. 1, Syuecheng Rd., Dashu Township, Kaohsiung County 840,Taiwan

Abstract: Fuzzy set theory has been used more and more frequently in intelligent systems because of its simplicity and similarity to human reasoning. It usually uses a fuzzy inference system to handle new cases for making decisions or controlling actions. In the past, Takagi and Sugeno proposed a well-known fuzzy model, namely TS fuzzy model, to improve the precision of inference results. In this paper, we try to automatically adjust the membership functions appropriate for the TS fuzzy model. A GA-based learning algorithm is thus proposed to achieve the purpose. The proposed approach considers the shapes of membership functions in fitness evaluation in addition to the accuracy. The experimental results show that the proposed approach can derive the membership functions in the Takagi-Sugeno system with low errors and good shapes.

Keywords: genetic algorithms; GAs; membership functions; fuzzy inference; Takagi-Sugeno fuzzy model; machine learning; fuzzy set theory; fuzzy logic; TS fuzzy model.

DOI: 10.1504/IJIIDS.2011.040087

International Journal of Intelligent Information and Database Systems, 2011 Vol.5 No.3, pp.229 - 245

Received: 01 Dec 2009
Accepted: 10 Aug 2010

Published online: 21 Oct 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article