模糊知識庫獲得、推理與驗證之研究A Study on Fuzzy Knowledge Acquisition, Reasoning and Verification
4
0
0
全文
(2)
(3) A Study on Fuzzy Knowledge Acquisition, Reasoning, and Verification
(4)
(5)
(6)
(7) !" #$%&'()*+
(8) ! "#$%&'()*+,-./0 1()23456789:;< 01=> ?@01ABCD EFGHIJK0123LMN OPQ01=> RS=> T U45VWXYZX[\]AB^L _ X=> X()* +X0123X()45 Abstract This project contains two main parts and can be explained as follows. In the first part of our work, we use the properties of learning and tolerance of neural networks to acquire knowledge. In the part of fuzzy knowledge reasoning and verification, we propose a new version of Petri nets to represent fuzzy rules in the knowledge base. In the proposed model, fuzzy inference can be performed by using a token passing scheme. In addition, by transforming fuzzy Petri nets into non-fuzzy Petri nets, the task of knowledge verification can be achieved. Keywords: Neural Networks, Petri Nets, Knowledge Acquisition, Fuzzy Inference, Knowledge Verification
(9) . O`abcS/dYefghij ?kl+abcSlmmno pmmqrLsabcStuvR Swxyz{|}~ Lu vab*+()zh 01$%`() RSC¡h¢£ ¤¥¦236A`ab()§ ¨©ª«¬®¯°013±²³ ´µ¶·¸$%O`|}~u0 1¹º¢»M013±l abcS()¼¤½´µ8¾¿ YÀ%p¿6ÁL.Â:67R SabcSpÃzh"#Ä´: Å OÆÇÈ4{ÈÉÊËÌÍ N 4C´ÎGÏÐ ()¼:
(10) O` ÑÒÓ z h XÔÕKÖ!"#´× »MØ¢01¦ ÙÚ Û&'()*+,-LÜÝ Þ()¼ßàáhâx()¼45 Au{ãLp»MäA B~abcSªåæ Ðç
(11) _è ()*+ÒéÑÓ ê ª å 2 3 w ¥ ë ( ) ¼ ì 4 íîÒÓ L°ï
(12) abcSÐ ðÃhqrñòL»M/ó ôõ´ö÷
(13) 89hâ&' 6AL .
(14) èø()*+ ùú()*+,- ûü :ýcSþ90McS ó!9þ Á
(15) : (9þÝM0cS ´µ ²Ï9M,-/ y
(16) p
(17) qr,-þ&ä h`¿´ýcSþÈÉ L896A 01 TÐ _ èøö9 !"$ %#$9£
(18) þAB§¨M 9 %ABóþ&'. ÎG#$C£
(19) (²TU r)*+,»M&'RS -.012±Ý&' /, -L;ABÎG0$6A`a bcSþ()¼¤½L êøRS1R2 3ÖA&'M ,-ÝÝ4U5
(20) ¬6},- O`u:7þ890$9 þ:æAB¥¤ þ;¥!<=>q}? O7@ ABCÁTUr) *+,»M` D,!5
(21) ¬ 6},-+UBEFL ä6Aæ¨ûG_ HI:_01 J#$<H:æ ABMAK|LMäý/ f JÐ:NN¿66 OfïNO f:P QRSTUV AB§¨9 þ »MhAB¥Ðþ()¼W §¨ý9þ LMAþ . X-XCDL HI|_1R2 3ÖA 8 7+Uþ:æABÁTUr)þ *+,Lþ8+UþAB ¥¤ ;¥%<=>q YOAB'&'ZRS[ \]!<=>q,-LO`7>q9
(22) ^`_`}»M Aó äBCTUr)þ*+,L}+: 8uö aøb cdþ±e. óf$g &' hZEJO` M; zh:i%ál+/ §¨ji kþÝ
(23) »M fzhji k%ál &' Ømn@ D,CBY opØm¿ +L HIç_v7 qrþ 0$9¿ *þAB7AB 012±6AJK2±LO`M ÝAB ^!&' /9TU: *,»M2±9ØmpB^ ÈÉ}L pG7È46A j%Lsh: ûG_ y = x2 sin( x1) + x1 cos( x2), 0 ≤ x1, x 2 ≤ π t7ûu:@L¨6A0 +Uþt7ûu|@L°0$9A Bûuç@Ló/ . .
(24) .
(25) Aóvw ëx y¢ÁTU zøzz{|{}zx ÕgÕ6~° aøb d896A A èzz ýy¢v´T U zøzè| ÕgÕ6~LOMl 6A+U
(26) EFL êø()23 /()236789:;<=> ?@()¼i ß01 => UÒTbíRÑ ÓVÓXÐ UTbRV"L skhGPAB_ r1: If (height is tall) then (weight is heavy). r2: If (height is tall) and (eating is much) then (weight is very heavy). r3: If (height is tall) then not (weight is balanced). r4: If (eating is much) and (sport is big-2) then not (weight is light). r5: If (eating is very much) then (weight is light). r6: If (eating is very much) and (sport is big-2) and (sleeping is not bad) then not (weight is not very heavy). r7: If (eating is much) and (sport is big-2) and (sleeping is not enough) then (weight is very heavy). r8: If (sleeping is not enough) then (weight is light). r9: If (sleeping is bad) then (weight is light). r10 If (weight is very heavy) then (height is very tall).. uç_0$+þAB ä UTbRV ?@+ØmûuLM . ;¥ u?@01AB YÞhß => UÒ TbíÓVÓXÐ UTbV" $ % û _ Þ È íÒÓ î îÒÓ"X- ÒÓ"X s ÒÑÓ"L. u_ UTbRV ?@()¼¹ MN ZTU0123^ DE Ó"?@ÈC/ FG023L¦:=> KL ëø()¼45. Z¦()¼?@A:ö UTbRV 089:01()¼456 A 6 A ì 4 £
(27) ( ) " û _ V W îÓ"XYZ ÑÒ "X[\ "]L 45HIX¨ûG_ èø l K Ô ÒÓ ÒÑÓ"ä01=> PQÐ j01=> Lu MHI+u »/Mæ÷01¸.
(28) “Inference and improper knowledge detection on fuzzy rules with enhanced high-level fuzzy petri nets,” in Proceedings of 1998 Conference of North American Fuzzy Information Processing Society (NAFIPS), pp. 135-139, Pensacola Beach, Florida, USA. -. C.-S. Ouyang, H.-L. Tsai, and S.-J. Lee, (1998,10), “Knowledge acquisition from input-output data by fuzzy neural systems,” in Proceedings of IEEE International Conference on Systems, Man, and Cybernetics, pp. 1928-1933, San Diego, California, USA.. u_äuPQÐ UTbV Øm &' }" LBnÏu9z0 1$%L êøä + U = > Ð 3 ÒÒ "C£
(29) () ¡«Ý$P¢L ëøäHI ê +U 3 35 Óí"ÙÚ453 ¶·% ÒÓîÒ£Ó "+(¤$ ¥£
(30) ()u¦/L8¨¹ä +UGØm_ è§ê §è§ë §ê§ë §{§x § {§| §¨§} ´ ÐYZABL ꧩ"§{§© §{§| §©§x § |§x ´ Ð V W L ° è§èz §ë§èz B´ Ð[\L ª«öHI ê989:7EFâx 6A«°ûüE¬l¿hâx u ®×^DL 9*
(31) ÐmC9¯°± ²`G©³±eºó_ - H.-T. Yang and S.-J. Lee, (1998,08). . [1]Y. Lin, G. Cunningham and S. Coggeshall, “Using fuzzy pa rtitions to create fuzzy systems from input-output data and set the initial weights in a fuzzy neural network,” IEEE Transactions on Fuzzy Systems, vol. 5, pp. 614-621, November 1997.. [2]Y. Lin and G. Cunningham, “A new approach to fuzzy-neur al system modeling,” IEEE Transactions on Fuzzy Systems, vol. 3, pp. 190-198, May 1995.. [3]C. F. Juang and C. J. Lin, “An on-line selfconstructing neural fuzzy inference network and its applications,” IEEE Transactions on Fuzzy Systems, vol. 6, pp.12-32, February 1998.. [4]K. S. Leung and Y. T. So, “Consistency checking for fuz zy expert systems,” International Journal of Approximate Reasoning, vol. 9, pp. 263-282, 1993.. [5]C. H. Wu and S. J. Lee, “Enhanced high-level Petri nets with multiple colors for knowledge validation/veri fication of rule-based expert systems,” IEEE Transactions on Systems, Man, and Cybernetics - Part B: Cybernetics, vol. 27, no. 5, pp. 760-773, 1997.. [6]S. M. Chen, “A new approach to handling fuz zy decision-making problems,” IEEE Transactions on Systems, Man, and Cybernetics - Part B: Cybernetics, vol. 18, no. 6, pp. 1012-1016, 1988..
(32)
相關文件
(2007), “Selecting Knowledge Management Strategies by Using the Analytic Network Process,” Expert Systems with Applications, Vol. (2004), “A Practical Approach to Fuzzy Utilities
and Liu, S.J., “Quantifying Benefits of Knowledge Management System: A Case Study of an Engineering Consulting Firm,” Proceedings of International Symposium on Automation and
Lange, “An Object-Oriented Design Method for Hypermedia Information Systems”, Proceedings of the Twenty-seventh annual Hawaii International Conference on System Sciences, 1994,
The scenarios fuzzy inference system is developed for effectively manage all the low-level sensors information and inductive high-level context scenarios based
Selcuk Candan, ”GMP: Distributed Geographic Multicast Routing in Wireless Sensor Networks,” IEEE International Conference on Distributed Computing Systems,
D.Wilcox, “A hidden Markov model framework for video segmentation using audio and image features,” in Proceedings of the 1998 IEEE Internation Conference on Acoustics, Speech,
Effectiveness of Simulation on Knowledge Acquisition, Knowledge Retention, and Self-Efficacy of Nursing Students in Jordan. Situated Learning Theory: The Key to Effective
Li, The application of Bayesian optimization and classifier systems in nurse scheduling, in: Proceedings of the 8th International Conference on Parallel Problem Solving
