Information Technology Reference
In-Depth Information
7.9 Concluding Remarks
In this chapter a similarity-driven rule base simplification method is presented.
This rule base simplification method serves two practical purposes: increase in
model transparency and decrease in computational cost. Furthermore, this method
can be combined with any data-driven automated fuzzy modelling procedure
together with genetic-algorithm-based fuzzy set tuning procedure to generate a
transparent yet accurate and compact fuzzy model. However, the efficiency of the
approach depends largely on three threshold parameter values which are currently
set by trial and error. Genetic algorithms or evolutionary computations, in general,
can possibly also be used here as a proper support tool to determine the optimum
values of these three threshold parameters.
References
[1]
Babuška R (1996) Fuzzy modelling and identification, Ph.D thesis, Delft University
of Technology, Delft, The Netherlands.
[2]
Barron AR (1993) Universal approximation bounds for superposition of a sigmoidal
function, IEEE Trans. Information Theory, vol. 39: 930-945.
[3]
Bezdek JC, Pal NR (1998) Some new indexes of cluster validity, IEEE Trans. on
System, Man and Cybernetics, 28(3):301-315
[4]
Chao CT, Chen YJ and Teng TT (1996) Simplification of fuzzy-neural systems using
similarity analysis, IEEE Trans. on SMC, Part-B: Cybernetics vol. 26: 344-354.
[5]
Cross V (1993) An analysis of fuzzy set aggregators and compatibility measures.
Ph.D. thesis, Wright State University, Ohio, USA.
[6]
Gath I, Geva AB (1989) Unsupervised optimal fuzzy clustering, IEEE Trans. Pattern
Analysis and Machine Intelligence. 11(7):773-781
[7]
Jain AK, Dubes RC (1988) Algorithms for clustering data, Prentice-Hall, Englewood
Cliffs, New Jersey.
[8]
Kaymak U, Babuška R (1995) Compatible cluster merging for fuzzy modelling, Proc.
of FUZZ-IEEE/IFES, Yokohama, Japan, 897-904
[9]
Klir GJ, Yuan B (1995) Fuzzy sets and fuzzy logic, Theory and Applications,
Prentice-Hall Inc., Upper Saddle River, New Jersey.
[10]
Palit AK and Babuška R (2001) Efficient training algorithm for Takagi-Sugeno type
neuro-fuzzy network, proc. of FUZZ-IEEE, Melbourne, Australia, vol. 3: 1367-1371.
[11]
Palit AK and Popoviü D (1999a) Fuzzy logic based automatic rule generation and
forecasting of time series, Proc. of FUZZ-IEEE, Seoul, Korea, vol. 1: 360-365
[12]
Palit AK and Popoviü D (1999b) Forecasting chaotic time series using neuro-fuzzy
approach, Proc. of IEEE-IJCNN. Washington DC, USA, vol. 3:1538-1543
[13]
Panchariya PC, Palit AK, Popovic D and Sharma AL, (2003a) Data driven simple
fuzzy rule generation algorithm for fuzzy modelling and identification, Proc. of First
Indian Internat. Conf. on AI (IICAI-03), Hyderbad, India, 1088-1097.
[14]
Panchariya PC, Palit AK, Popovic D and Sharma AL, (2003b) Simple fuzzy
modelling scheme for compact TS fuzzy model using real-coded Genetic algorithm,
Proc. of First Indian Internat. Conf. on AI (IICAI-03), Hyderbad, India, 1098-1107.
[15]
Panchariya PC, Palit AK, Popovic D and Sharma AL, (2004a) Nonlinear system
identification using Takagi-Sugeno type neuro-fuzzy model, Proc. of IEEE Intern.
Conf. on Intelligent Systems (IEEE-IS), Varna, Sofia, Bulgaria, vol. 1: 76-81.
Search WWH ::
Custom Search