Abstract
Regression refers to the problem of approximating measured data that are assumed to be produced by an underlying, possibly noisy function. However, in real applications the assumption that the data represent samples from one function is sometimes wrong. For instance, in process control different strategies might be used to achieve the same goal. Any regression model, trying to fit such data as good as possible, must fail, since it can only find an intermediate compromise between the different strategies by which the data were produced. To tackle this problem, an approach is proposed here to detect ambiguities in regression problems by selecting a subset of data from the total data set using TSK models, which work in parallel by sharing the data with each other in every step. The proposed approach is verified with artificial data, and finally utilised to real data of grinding, a manufacturing process used to generate smooth surfaces on work pieces.
Similar content being viewed by others
References
Goel P, Dedeoglu G, Roumeliotis SI, Sukhatme GS (2000) Fault detection and identification in a mobile robot using multiple model estimation and neural network In: IEEE international conference on robotics and automation (ICRA), San Fransisco
Maybeck PS, Pogoda DL (1989) Multiple model adaptive controller for the stol f-15 with sensor/actuator failures. In: Proceedings of the 20th conference on decision and control, pp 1566–1572
Quandt RE, Ramsey JB, (1978) Estimating mixtures of normal distributions and switching regressions. J Am Stat Assoc 73:730–752
Quandt RE (1972) A new approach to estimating switching regressions. J Am Stat Assoc 67:306–310
Ménard M (2001) Fuzzy clustering and switching regression models using ambiguity and distance rejects. Fuzzy Sets Syst 122(3):363–399
Hathaway RJ, Bezdek JC (1993) Switching regression models and fuzzy clustering. IEEE Trans Fuzzy Syst 1(3):195–203
Shen H, Yang J, Wang S (2004) Outlier detecting in fuzzy switching regression models. Lecture notes in computer science, Vol. 3192. Springer, Berlin Heidelberg New York, pp 208–215
Rao AV, Miller D, Rose K, Gersho A (1997) Mixture of experts regressions modeling by deterministic annealing. IEEE Trans Signal Process 45(11):2811–2820
Takagi T, Sugeno M (1985) Fuzzy identification of systems and its application to modelling and control. IEEE Trans Syst Man Cybern 15:116–132
Sugeno M, Kang GT (1988) Structure identification of fuzzy model. J Fuzzy Sets Syst 28:15–33
Klawonn F, Kruse R (1999) Techniques and applications of control systems based on knowledge-based interpolation. In: Leondes CT (eds) Fuzzy theory systems: Techniques and application. Academic, San Diego, pp 431–460
Kecman V, Pfeiffer B-M (1994) Exploiting the structural equivalence of learning fuzzy systems and radial basis function neural networks. In: Proceedings of European congress on intelligent techniques and soft computing (EUFIT’94), Achen, pp 58–66
Abonyi J, Babuska R, Chovan T, Szeifert F (2000) Incorporating prior knowledge in fuzzy c-regression models application to system identification. In: Proceedings of the intelligent systems in control and measurement symposium (INTCOM 2000), pp 99–110
Figueiredo M, Gomide F (1999) Design of fuzzy systems using neurofuzzy networks. IEEE Trans Neural Netw 10:815–827
Herrera F, Lozano M, Verdegay JL (1995) Tuning fuzzy logic controllers by genetic algorithms. Int J Approx Reason 12(3–4):299–315
Arslan A, Mehmet K (2001) Determination of fuzzy logic membership functions using genetic algorithms. J Fuzzy Sets Syst 118(2):297–306
Tsang ECC, Yeung DS (1999) Optimizing fuzzy knowledge base by genetic algorithms and neural networks. In: Proceedings of the IEEE international conference on systems, man, and cybernetics, Tokyo
Grauel A, Renners I, Ludwig LA (2000) Optimizing fuzzy classifiers by evolutionary algorithms. In: Proceedings of the IEEE 4th international conference on knowledge-based intelligent engineering systems and allied technologies
Bai A (2003) Determining fuzzy membership functions with Tabu search—an application to control. J Fuzzy Sets Syst 139(1):209–225
Simon D (2002) Training fuzzy systems with the extended Kalman filter. J Fuzzy Sets Syst 132(2):189–199
Cheng HD, Chen JR (1997) Automatically determine the membership function based on the maximum entropy principle. Inf Sci 96(3–4):163–182
King RI, Hahn RS (1986) Handbook of modern grinding technology. Chapman and Hall, London
Rao KGRK, Kuppuswamy G (1997) Fuzzy logic interface application to the electro-chemical grinding process analysis. In: Proceedings of 17th AIMTDR conference REC, Warangal, pp 329–334
Yang Z, Gao Y, Zhang D, Huang T (2003) A self-tuning based fuzzy-PID approach for grinding process control. Key Eng Materi 238–239:375–380
Tönshoff HK, Walter A (1994) Self-tuning fuzzy-controller for process control in internal grinding. J Fuzzy Sets Syst 63(3):359–373
Liao TW, Chen JL (1994) A neural network approach for grinding process: modeling and optimization. Int J Mach Tools Manuf 34(7):919–937
Satyanarayanan G, Lin IJ, Chen MK (1992) Neural network modelling and multi-objective optimisation of creep feed grinding of super alloys. Int J Prod Res 30(10):2421–2438
Nagasaka K, Ichihashi H, Leonard R (1995) Neuro-fuzzy GMDH and its application to modelling grinding characteristics. Int J Prod Res 33(5):1229–1240
Nandi AK, Banerjee MK (2005) FBF-NN-based modelling of cylindrical plunge grinding process using a GA J Mater Process Technol 162–163C:655–664
Lee CW, Shin YC (2003) Construction of fuzzy systems using least-squares method and genetic algorithm. J Fuzzy Sets Syst 137(3):297–323
Nandi AK, Pratihar DK (2004) Design of a genetic-fuzzy system to predict surface finish and power requirement in grinding. J Fuzzy Sets Syst 148(3):487–504
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Nandi, A.K., Klawonn, F. Detecting Ambiguities in Regression Problems using TSK Models. Soft Comput 11, 467–478 (2007). https://doi.org/10.1007/s00500-006-0110-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-006-0110-6