# Neuro-Fuzzy Expert Systems: Overview with a Case Study

## Abstract

Artificial neural networks or connectionist models xc1,2,3 are massively parallel interconnections of simple neurons that function as a collective system. They are designed perhaps as an attempt to emulate human performance and function (itinter- ligently). An advantage of neural nets lies in their high computation rate provided by massive parallelism, so that real-time processing of huge data sets becomes feasible with proper hardware. Information is encoded among the various connec- tion weights in a distributed manner. The multilayer perceptron (MLP) xc2 is a feed-forward neural network model consisting of multiple payers of simple, sigmoid processing elements (nodes) or neurons. After a lowermost input layer there are usually any number of intermediate of *hidden* layers followed by an output later at the top. The learning procedure has to determine the internal parameters of the hidden units on its knowledge of the inputs and desired outputs.

An expert system xc4,5 is a computer program that functions in a narrow domain dealing with specialized knowledge generally possessed by human experts. Such programs are very useful due to the usual shortage of qualified human experts in real life. The primary characteristics of an experts systems are a knowledge base designed with the help of a human expert, a narrow problem domain, and a performance on par with a human expert. The knowledge base is a problem-specific module containing information that controls inferencing. Traditional rule-based expert systems encoded this information as *If-Then* rules while the connectionist expert system xc6 uses the set of connection weights of a *trained* neural net model for this purpose. The inference engine is problem independent while the user interface links the external environment to the system. Connectionist experts systems are usually suitable in data-rich environment. They help in minimizing human interaction and associated inherent bias during the phase of knowledge base formation (which is time-consuming in case of traditional models) and also reduce the possibility of generating contradictory rules. The rule generation phase of such connectionist models are usually completely automated. An expert system is expected to be able to draw conclusions without seeing all possible external information. It should be capable of directing the acquisition of new information in an efficient manner and also be able to justify a conclusion reached.

## Keywords

Fuzzy Logic Expert System Connection Weight Certainty Factor Fuzzy Expert System## Preview

Unable to display preview. Download preview PDF.

## References

- [1]R. P. Lippmann, “An introduction to computing with neural nets,”
*IEEE Acoustics*,*Speech and Signal Processing Magazine*, vol. 61, pp. 4–22, 1987.Google Scholar - [2]D. B. Rumelhart and J. L. McClelland, eds.,
*Parallel Distributed Processing*. Vol. 1, Cambridge,MA: MIT, 1986.Google Scholar - [3]T. Kohonen,
*Self-Organization and Associative Memory*. Berlin: Springer-Verlag, 1989.Google Scholar - [4]E. Rich,
*Artficial Intelligence*. Singapore: McGraw Hill, Inc., 1986.Google Scholar - [5]K. Ng and B. Abramson, “Uncertainty management in expert systems,”
*IEEE Expert*, pp. 29–48, April 1990.Google Scholar - [6]S. I. Gallant, “Connectionist expert systems,”
*Communications of the ACM*, vol. 31, pp. 152–169, 1988.CrossRefGoogle Scholar - [7]G. J. Klir and T. Folger,
*Fuzzy Sets, Uncertainty and Information*. Reading, MA: Addison Wesley, 1989.Google Scholar - [8]S. K. Pal and D. Dutta Majumder,
*Fuzzy Mathematical Approach to Pattern Recognition*. New York: Wiley (Halsted Press), 1986.MATHGoogle Scholar - [9]J. C. Bezdek and S. K. Pal, eds.,
*Fuzzy Models for Pattern Recognition: Methods that Search for Structures in Data*. NY: IEEE Press, 1992.Google Scholar - [10]A. Kandel,
*Fuzzy Mathematical Techniques with Applications*. Reading, MA: Addison-Wesley, 1986.MATHGoogle Scholar - [11]H.-J. Zimmermann,
*Fuzzy Set Theory — and its Applications*. Boston: Kluwer, 1991.MATHGoogle Scholar - [12]L. A. Zadeh, “The role of fuzzy logic in the management of uncertainty in expert systems,”
*Fuzzy Sets and Systems*, vol. 11, pp. 199–227, 1983.MATHCrossRefGoogle Scholar - [13]Y. H. Pao,
*Adaptive Pattern Recognition and Neural Networks*. Reading,MA: Addison-Wesley, 1989.MATHGoogle Scholar - [14]B. Kosko,
*Neural Networks and Fuzzy Systems*. New Jersey: Prentice Hall, 1991.Google Scholar - [17]M. E. Cohen and D. L. Hudson, “Approaches to the handling of fuzzy input data in neural networks,” in
*Proceedings of 1st IEEE Conference on Fuzzy Systems*, (San Diego), pp. 93–100, 1992.Google Scholar - [18]S. K. Pal and S. Mitra, “Multi-layer perceptron, fuzzy sets and classification,”
*IEEE Transactions on Neural Networks*, vol. 3, no. 5, 1992.Google Scholar - [19]A. DiNola, S. Sessa, W. Pedrycz, and E. Sanchez,
*Fuzzy Relation Equations and their Applications to Knowledge Engineering*. Dordrecht,The Netherlands: Kluwer Academic Publishers, 1989.MATHGoogle Scholar - [20]L. Shastri, “A connectionist approach to knowledge representation and limited inference,”
*Cognitive Science*, vol. 12, no. 3, 1988.Google Scholar - [21]J. Pearl, “Fusion, propagation and structuring in belief networks,”
*Artificial Intelligence*, vol. 29, pp. 241–288, 1986.MATHCrossRefGoogle Scholar - [22]B. G. Buchanan and E. H. Shortliffe, eds.,
*Rule-based Expert Systems: The MYCIN Experiments of the Stanford Heuristic Programming Project*. Reading, MA: Addison-Wesley, 1984.Google Scholar - [23]G. Shafer,
*A Mathematical Theory of Evidence*. Princeton, NJ: Princeton University Press, 1976.MATHGoogle Scholar - [24]L. A. Zadeh, “Fuzzy sets as a basis for a theory of possibility,”
*Fuzzy Sets and Systems*, vol. 1, pp. 3–28, 1978.MATHCrossRefGoogle Scholar - [25]R. O. Duda, P. E. Hart, and N. J. Nilsson, “Subjective Bayesian methods for a rule-based inference system,” in
*Proceedings of the National Computer Conference*, (USA), pp. 1075–1082, 1976.Google Scholar - [26]I. B. Turksen and Z. Zhong, “An approximate analogical reasoning schema based on similarity measures and interval-valued fuzzy sets,”
*Fuzzy Sets and Systems*, vol. 34, pp. 323–346, 1990.CrossRefGoogle Scholar - [27]M. Ishizuka, K. S. Fu, and J. T. P. Yao, “Inference procedures under uncertainty for the problem-reduction method,”
*Information Sciences*, vol. 28, pp. 179–206, 1982.MATHCrossRefGoogle Scholar - [28]A. O. Esogbue and R. C. Elder, “Fuzzy sets and the modelling of physician decision processes, Part I: The initial interview — information gathering session,”
*Fuzzy Sets and Systems*, vol. 2, pp. 279–291, 1979.MATHCrossRefGoogle Scholar - [29]E. Sanchez and R. Bartolin, “Fuzzy inference and medical diagnosis, a case study,”
*Biomedical Fuzzy Systems Bulletin*, vol. 1, pp. 4–21, 1990.Google Scholar - [30]Z. A. Sosnowski, “An extension of Clips for processing fuzzy data,” Tech. Rep. NRCC Pub.No. 31506, Laboratory for Intelligent Systems, Division of Electrical Engineering, National Research Council, Ottawa, Canada, 1990.Google Scholar
- [31]D. C. Kuncicky and A. Kandel, “A fuzzy interpretation of neural networks,” in
*Proceedings of 3rd International Fuzzy Systems Association Congress, University of Washington*, (Seattle, Washington), pp. 113–116, 1989.Google Scholar - [32]H. Takagi, “Fusion technology of fuzzy theory and neural network — Survey and future directions,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 13–26, 1990.Google Scholar - [33]J. M. Keller and H. Tahani, “Implementation of conjunctive and disjunctive fuzzy logic rules with neural networks,”
*International Journal of Approximate Reasoning*, vol. 6, pp. 221–240, 1992.MATHCrossRefGoogle Scholar - [34]J. M. Keller, R. R. Yager, and H. Tahani, “Neural network implementation of fuzzy logic,”
*Fuzzy Sets and Systems*, vol. 45, pp. 1–12, 1992.MATHCrossRefGoogle Scholar - [35]H. Ishibuchi, H. Okada, and H. Tanaka, “Interpolation of fuzzy if-then rules by neural networks,” in
*Proceedings of 2nd International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 337–340, 1992.Google Scholar - [36]S. Nakanishi and T. Takagi, “Pattern recognition by neural networks and fuzzy inference,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 183–186, 1990.Google Scholar - [37]W.P. Zhuang, W. Z. Qiao, and T.H. Heng, “The truth-valued flow inference network,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 267–281, 1990.Google Scholar - [38]J. Yen, “The role of fuzzy logic in the control of neural networks,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 771–774, 1990.Google Scholar - [39]R. Masuoka, N. Watanabe, A. Kawamura, Y. Owada, and K. Asakawa, “Neurofuzzy system — fuzzy inference using a structured neural network,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 173–177, 1990.Google Scholar - [40]H. Okada, N. Watanabe, A. Kawamura, K. Asakawa, T. Taira, K. Ishida, T. Kaji, and M. Narita, “Knowledge implementation multilayer neural networks with fuzzy logic,” in
*Proceedings of 2nd International Conference on Fuzzy Logic and Neural Networks*,*Iizuka*, (Japan), pp. 99–102, 1992.Google Scholar - [41]R.R. Yager, “Implementing fuzzy logic controllers using a neural network framework,”
*Fuzzy Sets and Systems*, vol. 48, pp. 53–64, 1992.CrossRefGoogle Scholar - [42]H. Takagi and I. Hayashi, “Artificial neural network driven fuzzy reasoning,”
*International Journal of Approximate Reasoning*, vol. 5, pp. 191–212, 1991.MATHCrossRefGoogle Scholar - [43]A. Amano and T. Aritsuka, “On the use of neural networks and fuzzy logic in speech recognition,” in
*Proceedings of International Joint Conference on Neural Networks*, (Washington D.C.), pp. 301–305, 1989.Google Scholar - [44]H. Takahashi and H. Minami, “Subjective evaluation modelling using fuzzy logic and a neural network,” in
*Proceedings of 3rd International Fuzzy Systems Association Congress, University of Washington*, (Seattle, Washington), pp. 520–523, 1989.Google Scholar - [45]S. Horikawa, T. Furuhashi, and Y. Uchikawa, “A new type of fuzzy neural network for linguistic fuzzy modelling,” in
*Proceedings of 2nd International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 1053–1056, 1992.Google Scholar - [46]E. h. Zahzah, J. Desachy, and M. Zehana, “A fuzzy connectionist approach for a knowledge based image interpretation system,” in
*Proceedings of 2nd International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 1135–1138, 1992.Google Scholar - [47]C. C. Lee, “A self-learning rule-based controller employing approximate reasoning and neural network concepts,”
*International Journal of Intelligent Systems*, vol. 6, pp. 71–93, 1991.CrossRefGoogle Scholar - [48]D. G. Bounds, P. J. Lloyd, and B. G. Mathew, “A comparison of neural network and other pattern recognition approaches to the diagnosis of low back disorders,”
*Neural Networks*, vol. 3, pp. 583–591, 1990.CrossRefGoogle Scholar - [49]R. H. Silverman and A. S. Noetzel, “Image processing and pattern recognition in ultrasonograms by backpropagation,
*Neural Networks*, vol. 3, pp. 593–603, 1990.CrossRefGoogle Scholar - [50]H. Endo and H. Isshiki, “Application of “Associatron” to expert system,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks*,*Iizuka*, (Japan), pp. 751–754, 1990.Google Scholar - [51]A. F. Rocha, “K-Neural nets and expert reasoning,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 143–146, 1990.Google Scholar - [52]A.F. Rocha, M. Theoto, and P. Torasso, “Heuristic learning expert systems — general principles,” in
*Fuzzy Logic in Knowledge-Based Systems, Decision and Control*, (M. M. Gupta and T. Yamakawa, eds.), pp. 289–306, North-Holland: Elsevier Science Publishers B. V., 1988.Google Scholar - [53]R. J. Machado and A. F. Rocha, “A hybrid architecture for connectionist expert systems,” in
*Intelligent Hybrid Systems*, (A. Kandel and G. Langholz, eds.), CRC Press, 1992.Google Scholar - [54]Y. Hayashi, J. J. Buckley, and E. Czogala, “Approximation between fuzzy expert systems and neural networks,” in
*Proceedings of 2nd International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 135–139, 1992.Google Scholar - [55]H. F. Yin and P. Liang, “A connectionist incremental expert system combining production systems and associative memory,”
*International Journal of Pattern Recognition and Artificial Intelligence*, vol. 5, pp. 523–544, 1991.CrossRefGoogle Scholar - [56]K. Yoshida, Y. Hayashi, A. Imura, and N. Shimada, “Fuzzy neural expert system for diagnosing hepatobiliary disorders,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 539–543, 1990.Google Scholar - [57]Y. Hayashi, “A neural expert system with automated extraction of fuzzy if-then rules and its application to medical diagnosis,” in
*Advances in Neural Information Processing Systems*, (R. P. Lippmann, J. E. Moody, and D. S. Touretzky, eds.), pp. 578–584, Los Altos: Morgan Kaufmann, 1991.Google Scholar - [58]Y. Hayashi, “A neural expert system using fuzzy teaching input,” in
*Proceedings of 1st IEEE Conference on Fuzzy Systems*, (San Diego), pp. 485–491, 1992. 143Google Scholar - [59]D.L. Hudson, M. E. Cohen, and M.F. Anderson, “Use of neural network techniques in a medical expert system,” in
*Proceedings of the 3rd International Fuzzy Systems Association Congress, University of Washington*, (Seattle, Washington), pp. 476–479, 1989.Google Scholar - [60]D. L. Hudson, M. E. Cohen, and M. F. Anderson, “Use of neural network techniques in a medical expert system,”
*International Journal of Intelligent Systems*, vol. 6, pp. 213–223, 1991.CrossRefGoogle Scholar - [61]D. L. Hudson and M. E. Cohen, “Combination of rule-based and connectionist expert systems,”
*International Journal of Microcomputer Applications*, vol. 10, pp. 36–41, 1991.Google Scholar - [62]M. E. Cohen, D. L. Hudson, and M. F. Anderson, “A neural network learning algorithm with medical applications,” in
*Computer Applications in Medical Care*, (L. C. Kingsland, ed.), pp. 307–311, IEEE Computer Society Press, 1989.Google Scholar - [63]E. Sanchez, “Fuzzy connectionist expert systems,” in
*Proceedings of the 1990 International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 31–35, 1990.Google Scholar - [64]K. Saito and R. Nakano, “Medical diagnostic expert system based on PDP model,” in
*Proceedings of IEEE International Conference on Neural Networks*, (SanDiego, California), pp. I.255–I.262, 1988.Google Scholar - [65]S. G. Romaniuk and L. O. Hall, “Decision making on creditworthiness, using a fuzzy connectionist model,”
*Fuzzy Sets and Systems*, vol. 48, pp. 15–22, 1992.CrossRefGoogle Scholar - [66]S. K. Pal and P. K. Pramanik, “Fuzzy measures in determining seed points in clustering,”
*Pattern Recognition Letters*, vol. 4, pp. 159–164, 1986.CrossRefGoogle Scholar - [67]S. K. Pal and D. P. Mandal, “Linguistic recognition system based on approximate reasoning,”
*Information Sciences*, vol. 61, pp. 135–161, 1992.CrossRefGoogle Scholar - [68]S. Mitra and S. K. Pal, “Rule generation and inferencing with a layered fuzzy neural network,” in
*Proceedings of 2nd International Conference on Fuzzy Logic and Neural Networks, Iizuka*, (Japan), pp. 641–644, 1992.Google Scholar