臺灣博碩士論文加值系統

　　本論文針對分類之問題提出四種基於模糊系統的分類架構。首先，發展出一具有階層式的分類系統，該系統係藉由遺傳基因演算法進行修正。此模糊分類系統是由多個子模糊系統所建構，而所對應的模糊法則乃取決於資料的分佈情形。此外，基因演算法為搜尋輸入變數的組合及降低每一個子系統中的法則數目為主要目的。第二，建立具適應性的模糊分類系統。在模糊法則中結論部的輸出可被視為該條法則的信賴強度，本系統主要目的係提出一適應性模糊分類演算法去調整信賴強度。第三，運用透過遺傳基因演算法搜尋並決定模糊特徵擷取及模糊分類單元的歸屬函數分佈區間，其中模糊特徵擷取的機制乃是將原高維度資料映射到較低維的資料空間。最後，我們提出了一個整合式的模糊分類系統，其中包含了模糊特徵擷取及模糊分類單元，並透過遺傳基因演算法及適應性模糊系統分別進行修正，以獲得更高的鑑別率。為了呈現出所提出模糊分類系統的性能，我們採用三個具代表性的分類資料庫(Wine, Iris, Glass)來對所提出的模糊分類系統進行評估。

　The dissertation proposes four of fuzzy models to solve the classification problem. First, a hierarchical fuzzy classification model is set up, where several fuzzy subsystems are included. The IF-THEN rules of each subsystem are generated according to the distributions of the feature variables. Two genetic algorithms are utilized to determine the combination of the input features for each subsystem and reduce the number of rules in each fuzzy subsystem, respectively. Secondly, an adaptive fuzzy classification model is established. The confident value of the IF-THEN rule represents the rule weight interpreted as its confident strength. A novel adaptive modification algorithm is developed to tune the confident value of the fuzzy classification model. In the 3rd fuzzy model, the fuzzy feature extraction agent and the fuzzy classification unit are developed by using genetic algorithms to determine the distribution of the fuzzy sets for each original feature variable and the newborn feature variable, respectively. The fuzzy feature extraction agent can validly reduce the original feature dimensions. Finally, an integrated classification model is built, where the genetic algorithm and adaptive grade mechanism are propounded to tune the fuzzy feature extraction agent and fuzzy classification unit, respectively. In order to show the feasibility and efficiencies of the proposed fuzzy classification models, the well-known Wine, Iris and Glass databases are exploited to test the performances.

中文摘要 I
Abstract II
Acknowledgment III
Contents IV
List of Tables VI
List of Figures VIII
Glossary X
Chpater 1 Introduction 1
1.1 Preliminary 1
1.2 Classification model 5
1.3 Outline of the dissertation 8
Chpater 2 Hierarchical fuzzy classification model using GAs (Model 1) 11
2.1 Hierarchical fuzzy classification model 12
2.1.1 Generation of the fuzzy subsystem (Stage 1) 16
2.1.2 Determination of the classification unit (Stage 2) 19
2.2 Two types of GAs for the HFM 21
2.2.1 Type I GA: To search the combination of input feature variables for each subsystem 22
2.2.2 Type II GA: To reduce the IF-THEN rules of each subsystem 24
2.3 Simulation results 26
2.3.1 Training phase 27
2.3.2 Testing phase 29
2.4 Discussions 30
Chpater 3 An adaptive fuzzy classification model (Model 2) 31
3.1 Fuzzy classification model 32
3.1.1 Generation of fuzzy IF-THEN rules 33
3.1.2 Determination of classification unit 35
3.1.3 Example 36
3.2 The adaptive modification algorithm 41
3.3 Simulation results 45
3.3.1 Training phase 46
3.3.2 Testing phase 47
3.4 Discussions 48
Chpater 4 Two-stage fuzzy classification models (Models 3 and 4) 49
4.1. Introduction of two-Stage fuzzy classification models 51
4.2. FFEA and FCU 52
4.2.1 FFEA 53
4.2.2 FCUs 56
4.2.3 Example: Model 4 61
4.3 GA-based two-stage fuzzy classification model (Model 3) 65
4.4 Integated two-stage fuzzy classification model (Model 4) 67
4.4.1 GAs for the Model 4 68
4.4.2 AGM 71
4.5 Simulation results and discussions 72
4.6 Discussions 79
Chpater 5 Conclusions and future study 80
5.1 Conclusions 80
5.2 Future study 83
References 84
Autobiography 93

[1] J. Abonyi, J. A. Roubos and F. Szeifert, “Data-driven generation of compact, accurate,and linguistically sound fuzzy classifiers based on a decision-tree initialization,”International Journal of Approximate Reasoning, vol. 32, pp. 1-21, 2003.
[2] B. Alain, S. Atagiri and B.-H. Juang, “Pattern recognition using discriminative featureextraction,” IEEE Transactions on Signal processing, vol. 45, pp. 500-504, 1997.
[3] R. Babuska, J. Oosterhoff, A. Oudshoorn and P. M. Bruijn, “Fuzzy self-tuning PI control of pH in fermentation,” Engineering Applications of Artificial Intelligence,vol. 15, pp. 3-15, 2002.
[4] E. Baralis and P. Garza, “A lazy approach to pruning classification rules,” in Proc.IEEE Int. Conf. Data Mining, pp. 35-42, 2002.
[5] R. Battiti and A. M. Colla, “Democracy in neural nets: Voting schemes for
classification,” Neural Networks, vo1.7, pp. 691-707, 1994.
[6] N. Belacel and M. R. Boulassel, “Multicriteria fuzzy classification procedure procftn:methodology and medical application”, Fuzzy Sets and Systems, vol. 141, pp.203-217, 2004.
[7] J. C. Bezdek, Pattern recognition with fuzzy objective function algorithms, Plenum Press, 1981.
[8] M. Brameier and W. Banzhaf, “A comparison of linear genetic programming and
neural networks in medical data mining,” IEEE Transactions on Evolutionary
Computation, vol. 5, pp. 17-26, 2001.
[9] D. A. Chiang and N. P. Lin, “Partial correlation of fuzzy sets,” Fuzzy Sets Systems, vol. 110, pp. 209-215, 2000.
[10] S. B. Cho and J. H. Kim, “Combining multiple neural networks by fuzzy integral for robust classification,” IEEE Transactions on Systems, Man and Cybernetics, Part B, vo1.25, pp. 380-384, 1995.
[11] S.-B. Cho and J. H. Kim, “Multiple network fusion using fuzzy logic,” IEEE Transactions on Neural Networks, vo1.6, pp. 497-501, 1995.
[12] L. Corcoran and S. Sen, “Using real-valued genetic algorithms to evolve rule sets for classification,” in Proc. 1st IEEE Int. Conf. Evolutionary Computation, pp. 120-124, 1994.
[13] A. Devillez, M. Sayed-Mouchaweh and P. Billaudel, “A process monitoring module based on fuzzy logic and pattern recognition,” International Journal of Approximate Reasoning, vol. 37, pp. 43-70, 2004.
[14] M. Forina, Wine Recognition Database, Available via anonymous ftp from ics.uci.edu indirectory /pub/machine-learning-databases, 1991.
[15] S. Galichet, D. Dubois and H. Prade, “Categorizing classes of signals by means of fuzzy gradual rules,” in Proc. of 18th Int. Joint Conf. on Artificial Intelligence (IJCAI'03), pp. 1039-1044, 2003.
[16] J. Gama, “Functional trees for classification,” in Proc. 2001 IEEE Int. Conf. Data Mining (ICDM), pp. 147-154, 2001.
[17] S. Geva and J. Sitte, “A constructive method for multivariate function approximation by multilayer perceptions,” IEEE Transactions Neural Networks, vol. 3, pp. 621-624, 1992.
[18] J. R. Guerci, J. S. Goldstein and I. S. Reed, “Optimal and adaptive reduced-rank STAP,” IEEE Transactions on Aerospace and Electronic Systems, vol. 36, pp.647-663, 2000.
[19] N. R. Guo, T.-H. S. Li and C. P. Cheng, “GA Based Fuzzy Feature Extraction for Fuzzy Classification Problem,” in Proc. of International Conference on Computational Intelligence for Modeling Control and Automation, pp. 980-990, 2004.
[20] N. R. Guo, T.-H.S. Li and C.-L. Kuo, “Design of Hierarchical Fuzzy Model for Classification Problem using GAs,” Computers & Industrial Engineering, 2005, Accepted.
[21] N. R. Guo, T.-H. S. Li and C.-L. Kuo, “Hierarchical fuzzy model for classification
[22] L. K. Hansen and P. Salamon, “Neural network ensembles,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 12, pp. 993-1001, 1990.
[23] S. Hati and S. Sengupta, “Robust camera parameter estimation using genetic
algorithm,” Pattern Recognition Letters, vol. 22, pp. 289-298, 2001.
[24] R. Holve, “Rule generation for hierarchical fuzzy systems,” in Proc. of NAFIPS '97, pp. 444-449, 1997.
[25] Y. C. Hu, R. S. Chen and G. H. Tzeng, “Mining fuzzy association rules for
classification problems,” Computers and Industrial Engineering, vol. 43, pp. 735-750, 2002.
[26]Y. P. Kuo and T.-H. S. Li, “GA-based fuzzy PI/PD controller for automotive active suspension system,” IEEE Transactions on Industrial Electronics, vol. 46, pp.1051-1056, 1999.
[27]F. J. Iannarilli and P. A. Rubin, “Feature selection for multiclass discrimination via mixed-integer linear programming,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, pp. 779-783, 2003.
[28]H. Ishibuchi, H. E. Aguirre and K. S. Tanaka, “Multi-objective optimization with improved genetic algorithm,” in Proc. IEEE Int. Conf. SMC, pp. 3852-3857, 2000
[29]H. Ishibuchi, T. Murata and B. Turksen, “ Single-objective and two-objective genetic algorithms for selecting linguistic rules for pattern classification problems,” Fuzzy Sets Systems, vol. 89, pp. 135-150, 1997.
[30]H. Ishibuchi and T. Nakashima, “Performance evaluation of fuzzy classifier systems for multidimensional pattern classification problems,” IEEE Transactions on Systems,Man and Cybernetics, Part B, vol. 29, pp. 601-617, 1999.
[31]H. Ishibuchi and T. Nakashima, “Improving the performance of fuzzy classifier systems for pattern classification problems with continuous attributes,” IEEE Transactions Indus. Electronics vol. 46, pp. 1057-1068, 1999.
[32] H. Ishibuchi, T. Nakashima and T. Morisawa, “Voting in Fuzzy Rule-Based System for Pattern Classification Problems,” Fuzzy Sets and Systems, vol. 103, pp. 223-238,1999.
[33] H. Ishibuchi, K. Nozaki, N. Yamamoto and H. Tanaka, “Selecting fuzzy if-then rules for classification problems using genetic algorithms,” IEEE Transactions on Fuzzy Systems, vol. 3, pp. 260-270, 1995.
[34] E. R. Kandel, J. H. Schwartz and T. M. Jesse, Principles of Neural Science, New York,Elsevier, 1991.
[35] N. B. Karayiannis, “Generalized fuzzy c-means algorithm,” Journal of Intelligent and Fuzzy Systems, vol. 8, pp. 63-81, 2000.
[36] N. K. Kasabov, “Learning fuzzy rules and approximate reasoning in fuzzy neural networks and hybrid system,” Fuzzy Sets Systems, vol. 82, pp. 135-149, 1996.
[37] J. K. Kishore, L. M. Patnaik, V. Mani and V. K. Agrawal, “Application of genetic programming for multicategory pattern classification,” IEEE Transactions on Evolutionary computation, vol. 4, pp. 242-257, 2000.
[38] M. Kudo and J. Sklansky, “Comparison of algorithms that select features for pattern classifiers,” Pattern Recognition, vol. 33, pp. 25-41, 2000.
[39] N. Kwak and C. -H. Choi, “Feature extraction based on ica for binary classification problems,” IEEE Transactions on Knowledge and Data Engineering, vol. 15, pp.1374-1388, 2003.
[40] H. M. Lee, C. M. Chen and J. M. Chen, “An efficient fuzzy classier with feature selection based on fuzzy entropy,” IEEE Transactions on Systems, Man and Cybernetics, Part B, vol. 31, pp. 426-432, 1999.
[41] J. H. Lee, S. J. Yu and S. C. Park, “Design of intelligent data sampling methodology based on data mining,” IEEE Transactions on Robotics and Automation, vol. 17, pp. 637-649, 2001.
[42] Y. Leung, J.-H. Ma and W.-X. Zhang, “A new method for mining regression classes in large data sets,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, pp. 5-21, 2001.
[43] R.-P. Li, M. Mukaidono and I. B. Turksen, “A fuzzy neural network for pattern classification and feature selection,” Fuzzy Sets and Systems, vol. 130, pp. 101-108, 2002.
[44] T.-H. S. Li, N. R. Guo and C.-L. Kuo, “Design of adaptive fuzzy model for
classification problem,” Engineering Applications of Artificial Intelligence, vol. 18, pp. 297-306, 2005.
[45] T.-H. S. Li and M.–Y. Shieh, “Design of a GA-based fuzzy PID controller for non-minimum phase systems,” Fuzzy Sets Systems, vol. 11, pp. 183-197, 2000.
[46] D. A. Linkens and M. Y. Chen, “Hierarchical fuzzy clustering based on
self-organizing networks,” in Proc. IEEE Int. Conf. Fuzzy Systems, pp. 1406-1410, 1998.
[47] J. A. Lozano and P. Larranaga, “Applying genetic algorithms to search for the best hierarchical clustering of a dataset,” Pattern Recognition Letters, vol. 20, pp.911-918, 1999.
[48] H.-M. Mads, M. W. Pedersen, L. K. Hansen and J. Larsen, “Design and evaluation of neural classifiers,” in Proc. 6th Int. Conf. Neural Networks for Signal Processing, pp.223-232, 1996.
[49] S. Matsushita, T. Furuhashi, H. Tsutsui and Y. Uchikawa, “Efficient search for fuzzy models using genetic algorithm,” Info. Sciences, vol. 110, pp. 41-50, 1998.
[50] G. J. McLachlan, Discriminant Analysis and Statistical Pattern Recognition, Wiley, 1992.
[51] W. Melek, A. Goldenberg and M. R. Emami, “A fuzzy noise-rejection data
partitioning algorithm,” International Journal of Approximate Reasoning, vol. 38, pp.1-17, 2005.
[52] S. Mondal and M. Maiti, “Multi-item fuzzy EOQ models using genetic algorithm,”Computers and Industrial Engineering, vol. 44, pp. 105-117, 2003.
[53] T. Nakasbima and H. Ishibuchi, “Supervised and unsupervised fuzzy discrimination of Continuous Attributes for Pattern Classification Problems,” in Proc. of Knowledge-Based Intelligent Information Engineering Systems & Allied
Technologies, pp. 32-36, 2001.
[54] B. Novakovic, D. Scap and D. Novakovic, “An analytic approach to fuzzy robot control synthesis,” Engineering Applications of Artificial Intelligence, vol. 13, pp.71-83, 2000.
[55] K. Nozaki, H. Ishibuchi and H. Tanaka, “Adaptive fuzzy rule-based classification systems,” IEEE Transactions Fuzzy System, vol. 4, pp. 238-250, 1996.
[56] M. Padmanabhan and L. R. Bahl, “Partitioning the feature space of a classifier with linear hyperplanes,” in Proc. IEEE Transactions Speech and Audio, pp. 282-288, 1999.
[57] R. Parekh, J. Yang and V. Honavar, “Constructive neural-network learning algorithms for pattern classification,” IEEE Transactions on Neural Networks, vol. 11, pp. 436-451, 2000.
[58] J. R. Quinlan, “Induction of decision trees,” Machine Learning, vol. 1 pp. 81-106,1986.
[59] V. Ravi and J. Zimmermann, “Fuzzy rule based classification with feature selector and modified threshold accepting,” European Journal of Operational Research, vol. 123, pp. 16-28, 2000.
[60] K. S. Ray and J. Ghoshal, “Neuro-genetic approach to multidimensional fuzzy reasoning for pattern classification,” Fuzzy Sets and Systems, vol. 112, pp. 449-483,2000.
[61] M. L. Raymer, W. F. Punch, E. D. Gooidman, L. A. Kuhn and A. K. Jain,
“Dimensionality reduction using genetic algorithms,” IEEE Transactions on
Evolutionary Computation, vol. 4, pp. 164-171, 2000.
[62] D. T. Redden, “Further examination of fuzzy linear regression,” Fuzzy Sets Systems, vol. 79, pp. 203-211, 1996.
[63] F. Ricci and P. Avesani, “Data compression and local metrics for nearest neighbor classification,” IEEE Transactions on Pattern analysis and machine intelligence, vol. 21, pp. 380-384, 1999.
[64] B. Ripley, Pattern Recognition and Neural Networks, Cambridge Univ. Press, 1996.
[65] A. Ruiz and E. L. Pedro, “Nonlinear kernel-based statistical pattern analysis,” IEEE Transactions on Neural Networks, vol. 12, pp. 16-32, 2001.
[66] D. A. Savic and W. Pedrycz, “Evaluation of fuzzy linear regression model,” Fuzzy Sets Systems, vol. 39, pp. 51-63, 1991.
[67] M. Setnes and H. Roubos, “GA-fuzzy modeling and classification: complexity and performance,” IEEE Transactions on Fuzzy Systems, vol. 8, pp. 509-522, 2000.
[68] H. J. Song and H. J. Kim, “Using genetic algorithms to work out index configuration for the class-hierarchy indexing in object databases,” Info. and Software Tech., vol.42, pp. 731-741, 2000.
[69] A. N. Srivastava, R. Su and A. S. Weigend, “Data mining for features using
scale-sensitive gated experts,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21, pp. 1268-1279, 1999.
[70] M. Stenes and H. Roubos, “GA-fuzzy modeling and classification: complexity and performance,” IEEE Transactions on Fuzzy Systems, vol. 8, pp. 509-522, 2000.
[71] D. Suc, D. Vladusic and I. Bratko, “Qualitatively faithful quantitative prediction,” in Proc. of 18th Int. Joint Conf. on Artificial Intelligence (IJCAI'03), pp. 1052-1060,2003.
[72] L. -X. Sun, K. Danzer and G. Thiel, “Classification of wine samples by means of artificial neural networks and discrimination analytical methods,” J. Anal Chem., vol. 359, pp. 143-149, 1997.
[73] H. Tanaka, “Fuzzy data analysis by possibility linear models,” Fuzzy Sets Systems,vol. 24, pp. 363-375, 1987.
[74] W. Tong and T.-H. S. Li, “Realization of two-dimensional target tracking problem via autonomous mobile robots using fuzzy sliding mode control,” in Proc. of 24th Int.Conf, Aachen, pp. 1158-1163, 1998.
[75] A. Vernet and G. A. Kopp, “Classification of turbulent flow patterns with fuzzy clustering,” Engineering Applications of Artificial Intelligence, vol. 15, pp. 315-326, 2002.
[76] X. Wang and J. Hong, “Learning optimization in simplifying fuzzy rules,” Fuzzy Sets and Systems, vol. 106, pp. 349-356, 1999.
[77] S. M. Weiss and C. A. Kulikowski, Computer systems that learn, classification and prediction methods from statistics machine learning and expert systems, California Morgan Kaufmann Publishers, 1991.
[78] T. P. Wu and S. M. Chen, “A new method for constructing membership functions and fuzzy rules from training examples,” IEEE Transactions on Systems, Man and Cybernetics, Part B, vol. 29, pp. 334-347, 1999.
[79] N. Xiong, L. Litz and H. Ressom, “Learning premise of fuzzy rules for knowledge acquisition in classification problems,” Knowledge and Information Systems, vol. 4, pp. 96-111, 2002.
[80]L. Xu, A. Krzyzak and C. Y. Snen, “Methods of combining multiple classifiers and their applications to handwriting recognition,” IEEE Transactions on Systems, Man and Cybernetics, Part B, vol. 22, pp. 418-435, 1992.
[81] J. Yang and V. Honavar, “Feature subset selection using a genetic algorithm,” IEEE Intelligent, vol. 13, pp. 44-49, 1998
[82] N. Ye and X. Li, “A scalable incremental learning algorithm for classification problems,” Computers and Industrial Engineering, vol. 43, pp. 677-692, 2002.
[83] D. Yeung, “Constructive neural networks as estimators of bayesian discriminant functions,” Pattern Recognition, vol. 26, pp. 189-204, 1993.
[84] Y. Yuan and H. Zhuang, “A genetic algorithm for generating fuzzy classification rules,” Fuzzy Sets Systems, vol. 84, pp. 1-19, 1996.
[85] Y. Yuan and H. Zhuang, “A genetic algorithm for generating fuzzy classification rules,” Fuzzy Sets and Systems, vol. 84, pp. 1-19, 1996.
[86] H.-J. Zimmermann, Fuzzy set theory and its application. Kluwer academi, 1991.