Expectation reduction-based fuzzy nonlinear regression

doi:10.3969/j.issn.1000-1565.2018.01.014

Abstract

Abstract: Inspired by the idea of reduction in 2-fuzzy theory, an expectation reduction-based fuzzy nonlinear regression model is proposed to deal with triangular fuzzy number inputs and outputs.In this model the triangular fuzzy inputs and outputs are firstly replaced by their expectations, then the data can be trained with classical random weight network.Finally the crisp real outputs are recover to triangular fuzzy outputs by using a width matrix of original target triangular fuzzy outputs.The experiment results show that the proposed model gives higher learning accuracy and better generalization.

Key words: expectation reduction, triangular fuzzy number, random weight network, fuzzy nonlinear regression

CLC Number:

TP183

WANG Xizhao, ZHAO Shixin. Expectation reduction-based fuzzy nonlinear regression[J]. Journal of Hebei University (Natural Science Edition), 2018, 38(1): 89-98.

References

[1] WANG H F, TSAUT R C.Insight of a fuzzy regression model[J].Fuzzy Sets and Systems, 2000, 112(3): 355-369.DOI: 10.1016/S0165-0114(97)00375-8.
[2] KIMURA D, NII M, YAMAGUCHI T, et al.Fuzzy nonlinear regression analysis using fuzzified neural networks for fault diagnosis of chemical plants[J].Journal of Advanced Computational Intelligence and Intelligent Informatics, 2011, 15(3): 336-344.DOI: 10.20965/jaciii.2011.p0336.
[3] GONZALEZ-GONZALEZ D S, PRAGA-ALEJO R J, CAN-SIFUENTES M, et al.A non-linear fuzzy regression for estimating reliability in a degradation process[J].Applied Soft computing, 2014, 16: 137-147.DOI: 10.1016/j.asoc.2013.11.019.
[4] LIU Y Y, ZHOU J, CHEN Y Z.Using fuzzy non-linear regression to identify the degree of compensation among customer requirements in QFD[J].Neurocomputing, 2014, 142: 115-124.DOI: 10.1016/j.neucom.2014.01.053.
[5] WANG X Z.Learning from big data with uncertainty[J].Journal of Intelligent & Fuzzy Systems, 2015, 28(5): 2329-2330.
[6] ISHIBUCHI H, TANAKA H.Fuzzy regression analysis using neural networks[J].Fuzzy Sets and Systems, 1992,50(3):257-265.DOI: 10.1016/0165-0114(92)90224-R.
[7] RUMELHART D E, HINTON G E, WILLIAMS R J.Learning representations by back-propagating errors[J].Nature, 1986, 323: 533-536.DOI: 10.1038/323533a0.
[8] ISHIBUCHI H, TANAKA H.An architecture of neural networks with interval weights and its application to fuzzy regression analysis[J].Fuzzy Sets and Systems, 1993, 57(1): 27-39.DOI: 10.1016/0165-0114(93)90118-2.
[9] ISHIBUCHI H, KWON K, TANAKA H.A learning algorithm of fuzzy neural network with triangular fuzzy weights[J].Fuzzy Sets and Systems, 1995, 71(3): 277-293.DOI: 10.1016/0165-0114(94)00281-B.
[10] CHENG C B, LEE E S.Fuzzy regression with radial basis function network[J].Fuzzy Sets and Systems, 2001, 119(2): 291-301.DOI: 10.1016/S0165-0114(99)00098-6.
[11] ZHANG D, DENG L F, CAI K Y, et al.Fuzzy nonlinear regression with fuzzified radial basis function network[J].IEEE Transactions on Fuzzy Systems, 2005, 13(6): 742-760.DOI: 10.1109/TFUZZ.2005.859307.
[12] HE Y L, WANG X Z, HUANG Z X.Fuzzy nonlinear regression analysis using a random weight network[J].Information Sciences, 2016,364-365: 222-240.DOI: 10.1016/j.ins.2016.01.037.
[13] SCHMIDT W F, KRAAIJVELD M A, DUIN RPW.Feedforward neural networks with random weights[C].Proceedings of 11th IAPR International Conference on Pattern Recognition, Conference B: Pattern Recognition Methodology and Systems, II, Netherlands: IEEE, 1992,1-4.DOI: 10.1109/ICPR.1992.201708.
[14] PAO Y, TAKEFUJI Y.Functional-link computing: Theory, system architecture, and functionalities[J].Computer, 1992, 3: 76-79.DOI: 10.1109/2.144401.
[15] HUANG G B, SIEW C.Extreme learning machine: RBF network case[Z]. The 8th International Conference on Control, Automation, Robotics and Vision, Xiamen, 2004.
[16] HUANG G B, ZHU Q Y, SIEW C.Extreme learning machine: theory and application[J].Neurocomputing, 2006,70(1-3):489-501.DOI: 10.1016/j.neucom.2005.12.126.
[17] CAO F L, TAN Y P, CAI M M.Sparse algorithms of random weight networks and applications[J].Expert Systems with Applications, 2014, 41(5): 2457-2462.DOI: 10.1016/j.eswa.2013.09.045.
[18] CAO F L, YE H L, WANG D H.A probabilistic learning algorithm for robust modeling using neural networks with random weights[J].Information Sciences, 2015, 313: 62-78.DOI: 10.1016/j.ins.2015.03.039.
[19] 刘彦奎,陈艳菊,刘颖,等.模糊优化方法与应用[M].北京:科学出版社,2013.
[20] LIU B D, LIU Y K.Expected value of fuzzy variable and fuzzy expected value models[J].IEEE transactions on Fuzzy Systems,2002, 10(4): 445-450.
[21] CAMPOS L, GONZALEZ A.A subjective approach for ranking fuzzy numbers[J].Fuzzy Sets and Systems, 1989,29(2):145-153.
[22] DUBOIS D, PRADE H.The mean value of a Fuzzy number[J].Fuzzy Sets and Systems, 1987, 24(3): 279-300.DOI: 10.1016/0165-0114(87)90028-5.
[23] GONZALEZ A.A study of the ranking function approach through mean values[J].Fuzzy Sets and Systems, 1990, 35: 29-41.DOI: 10.1016/0165-0114(90)90016-Y.
[24] HEILPERN S.The expected value of a fuzzy number[J].Fuzzy Sets and Systems, 1992, 47(1): 81-86.DOI: 10.1016/0165-0114(92)90062-9.
[25] YAGER R.A procedure for ordering fuzzy subsets of the unit interval[J].Information Sciences,1981, 24(1): 143-161.DOI: 10.1016/0020-0255(81)90017-7.