臺灣博碩士論文加值系統

在發展推測系統時，會希望將專家的知識加入系統中。但冗長與專業的建構過程，使得專家系統無法深入各階層的應用。專家系統的建立，必須有固定的輸入資訊與穩定的推論個體。然而推論個體的不確定，建立系統時期或推論時期可能不斷地有新的推論個體加入或離開，使得推論個體推論時的動態變化，常影響到系統的建立。另一方面因為在建立系統時，所有需要的環境數據可由歷史資料得之，所以當系統建立時，採集的環境數據為環境歷史資料，因而能充分了解環境資訊。然而建立系統時期，歷史資料雖是齊全的，但是在推論(Inference)時，環境數據可能因為發佈的時間而無法取得。在推論的過程中，理論上環境狀態的資訊一定是非常明確的，但是實際上在推論時不一定能將所有資訊收集完畢。這些環境數據可能因為要作資料統計或尚無充分資訊而無法馬上得之。本論文提出動態決策資訊模型(Dynamic Decision support system model)即以專家建立初始模組、在推論時可自動補足不足資訊與滿足決策資訊變動性之決策資訊模型，以最少量之資訊提供決策，幫助建立一個完善的決策模型。

This dissertation proposes a method that contains the knowledge of experts in the Inference system while developing expert system. However, the process of establishing are so complicated, redundant, and professional such that the expert system can’t be applied into all levels. Furthermore, expert system must build on stationary input information and stable inferred individual. But, the unsure of the inferring individual, continuously entering or exiting by the new inferring individuals during system built or inferred time cause the inferred individual dynamic change, and then affect the establishing of the system.On the other hand, all the required information of environmental factor may get from historical data. The information of environmental factor can be sufficiently understood because the collected information is environment historical data while the system is building. However, during the time of system building, although the historical data may be complete, the information of environmental factor may not be gained during inferring because of the publish time. During the inferring process, theoretically the information of environmental factor should be very clear. In fact, we may not completely collect the data. The information of environmental factor may not be collected immediately because they are either used for data statistics or still incomplete.This paper proposes a Dynamic Decision Support System Model, which is an expert-based decision support system model that establishes initial model, automatically complement the incomplete information and satisfy the dynamic decision information during inferring, and provides decision and help with the minimum information to build a perfect decision system model.

1 Introduction
1.1 Preface 1
1.2 Background & Motivation 3
1.3 The Purpose of the Study 10
1.4 Organization of the Dissertation 14
2 Background Knowledge
2.1 The Browse of the Background
Knowledge 16
2.2 The Expert System 17
2.3 Neural Net Theory 26
2.4 The Establishment of Neural
Network 35
2.5 The Types of Machine Learning 38
2.6 The Learning Method of Neural
Network Theory 41
2.7 The Integration of Fuzzy Theory and
Neural Network Theory 45
2.8 The Establishment and Learning
of the Dynamic Decision Information
Model 48
3 The Establishment of Dynamic Decision Information Model
3.1 The Framework of the Prediction Model 51
3.2 The Framework of Dynamic
Decision Information Model 57
3.3 The Composition of the Model 59
3.4 The Insufficiency of Information
Incomplete 62
3.5 The Dynamic Change of a
Decision Entity during Inference 69
3.6 The Stages for Establishing Model 73
3.7 The Procedures of Establishment 75
3.8 The Procedures of Learning 77
3.9 The Method of Inference 80
4 The Result of Experiment
4.1 The Inference Model of the Behavior
of Fund Customer 83
4.2 Description of the Problems 85
4.3 The Establishment of the Model 87
4.4 Data Preparation 91
4.5 The Expert Define the Relation 97
4.6 Result Analysis 98
5 Conclusions
5.1 The Achievement 107
5.2 Integrity Other Method 108
5.3 The Other Applications 112
5.4 The Study in the Future 117

Reference
[1] Abecker * aabecker@dfki.uni-kl.de, A. Bernardi bernardi@dfki.uni-kl.de, H. Maus maus@dfki.uni-kl.de, M. Sintek sintek@dfki.uni-kl.de and C. Wenzel wenzel@dfki.uni-kl.de , “Information supply for business processes: coupling workflow with document analysis and information retrieval” , Knowledge-based Systems, Vol. 13 (5) (2000) pp. 271-284 © 2000 Elsevier Science B.V. All rights reserved. PII: S0950-7051(00)00087-3
[2] Albrecht, J., Lehner, W., Teschke, M. and Kirsche, T. “Building a real data warehouse for market research” Database and Expert Systems Applications, 1997. Proceedings., Eighth International Workshop on , 1997 , Page(s): 651 —656
[3] Amitava Dutta a * adutta@gmu.edu and Shyam Heda b , “Information systems architecture to support managed care business processes” , Decision Support Systems, Vol. 30 (2) (2000) pp. 217-225 © 2000 Elsevier Science B.V. All rights reserved. PII: S0167-9236(00)00098-1
[4] C. J. Matheus, P. K. Chan and G. Piatetsky-Shapiro, “systems for Knowledge Discovery in Databases,” IEEE trans. On Knowledge and Data Engineering, vol. 5. No. 6, pp. 903-913,1993.
[5] C. Olaru, L. Wehenkel, "Data mining", IEEE Computer Applications in Power, Volume: 12 3, pp.19-25, July 1999.
[6] C.W. Holsapple a * cwhols00@ukcc.uky.edu and K.D. Joshi1 b joshi@wsu.edu, “Organizational knowledge resources”, Decision Support Systems, Vol. 31 (1) (2001) pp. 39-54 © 2001 Elsevier Science B.V. All rights reserved. PII: S0167-9236(00)00118-4
[7] C.W. Holsapple, ” Knowledge management support of decision making”, Decision Support Systems, Vol. 31 (1) (2001) pp. 1-3 © 2001 Elsevier Science B.V. All rights reserved. PII: S0167-9236(00)00115-9
[8] Chung-Kwan Shin, Ui Tak Yun, Huy Kang Kim and Sang Chan Park, “Hybrid approach of neural network and memory-based learning of data mining” , IEEE Transactions on Neural Networks, Vol. 11 (3) (2000) pp. 637-646 © May 2000 IEEE. All rights reserved
[9] D. O. Hebb, The Organization of Behavior. John Wiley & Sons, New York, 1949.
[10] Don Peppers Martha Rogers The One to One Future :Building Relationships One Customer at a time 1996.
[11] E. Tsui a * eric_tsui@csc.co.nz, B.J. Garner b and S. Staab c , “The role of artificial intelligence in knowledge management” , Knowledge-based Systems, Vol. 13 (5) (2000) pp. 235-239 © 2000 Elsevier Science B.V. All rights reserved. PII: S0950-7051(00)00093-9
[12] Frada Burstein1 a *, Tung Bui2 and David Arnott3,” Introduction Decision support in the new millennium“, Decision Support Systems, Vol. 31 (2) (2001) pp. 163-164 © 2001 Elsevier Science B.V. All rights reserved. PII: S0167-9236(00)00128-7
[13] G. G. Piatetsky-Shapiro , “An Overview of Knowledge Discovery in Databases: Recent Progress and Challenges,” Proceedings of the International Workshop on Rough Sets and Knowledge Discovery, Banff, Alberta, Canada, 1993.
[14] Gregory Piatetsky-Shapiro, “Data-mining industry coming of age” , IEEE Expert, Vol. 14 (6) (1999) pp. 32-34 © 1999 IEEE. All rights reserved
[15] Hiroshi Tsukimoto , “Extracting rules from trained neural networks” , IEEE Transactions on Neural Networks, Vol. 11 (2) (2000) pp. 377-389 © 2000 IEEE. All rights reserved
[16] Int. J. Intell. Syst. (Special Issue on Knowledge Discovery in Data- and Knowledge Bases), vol. 7, no. 7, 1992.
[17] J. A. Feldman and D. H. Ballard, “Connectionist models and their properties,” Cognitive Science, vol. 6, pp. 205-254, 1982.
[18] J. J. Hopfield and D. W. Tank, ``Computing with neural circuits: A model,'' Science, vol. 233, pp. 625-633, August 1986.
[19] J. J. Hopfield, ``Neurons with graded response have collective computational properties like those of two-state neurons,'' Proc. Natl. Acad. Sci. USA, vol. 81, pp. 3088-3092, May 1984.
[20] J. J. Hopfield, ``Neural networks and physical systems with emergent collective computational abilities,'' Proc. Natl. Acad. Sci. USA, vol. 79, pp. 2554-2558, April 1982.
[21] J. J. Hopfield, ``Neurons with graded response have collective computational properties like those of two-state neurons,'' Proc. Natl. Acad. Sci. USA, vol. 81, pp. 3088-3092, May 1984.
[22] James Y.L. Thong * a jthong@ust.hk , Patrick Y.K. Chau b pchau@business.hku.hk and Kar Yan Tam a kytam@ust.hk,” Decision-making and e-commerce systems”, Decision Support Systems, Vol. 32 (1) (2001) pp. 1-2 © 2001 Elsevier Science B.V. All rights reserved. PII: S0167-9236(01)00096-3
[23] Jennifer Kreiea * jkreie@nmsu.edu, Timothy Paul Cronan1 b cronan@comp.uark.edu, John Pendley2 c pendley@email.uah.edu and Janet S. Renwick3 b renwick@comp.uark.edu , “Applications development by end-users: can quality be improved?” , Decision Support Systems, Vol. 29 (2) (2000) pp. 143-152 © 2000 Elsevier Science B.V. All rights reserved. PII: S0167-9236(00)00068-3
[24] John Zeleznikow a * johnz@cs.latrobe.edu.au and James R. Nolan b JNolan@siena.edu, “Using soft computing to build real world intelligent decision support systems in uncertain domains” , Decision Support Systems, Vol. 31 (2) (2001) pp. 263-285 © 2001 Elsevier Science B.V. All rights reserved. PII: S0167-9236(00)00135-4
[25] Kumar Chellapilla and David B. Fogel , “Evolving neural networks to play checkers without relying on expert knowledge” , IEEE Transactions on Neural Networks, Vol. 10 (6) (1999) pp. 1382-1391 © 1999 IEEE. All rights reserved
[26] Li Min Fu and Edward H. Shortliffe, “Application of certainty factors to neural computing for rule discovery” , IEEE Transactions on Neural Networks, Vol. 11 (3) (2000) pp. 647-657 © May 2000 IEEE. All rights reserved
[27] M. Smith, Neural Networks for Statistical Modeling. Van Nostrand Reinhold, 1993.
[28] M.J.A. Berry & G. Linoff : Data Mining Techniques, JohnWiley & Sons, Inc. 1997.
[29] Mauricio Figueiredo and Fernando Gomide , “Design of fuzzy systems using neurofuzzy networks” , IEEE Transactions on Neural Networks, Vol. 10 (4) (1999) pp. 815-827 © 1999 IEEE. All rights reserved
[30] P.G. Chander, T. Radhakrishnan and R. Shinghal, “Design schemes for rule-based systems” , International Journal of Expert Systems, Vol. 10 (1) (1997) pp. 1-36 © 1997 Elsevier Science. All rights reserved.
[31] P.K. Kannan a pkannan@rhsmith.umd.edu and H. Raghav Rao b mgmtrao@buffalo.edu, “Introduction to the special issue: decision support issues in customer relationship management and interactive marketing for e-commerce”, Decision Support Systems, Vol. 32 (2) (2001) pp. 83-84 © 2001 Elsevier Science B.V. All rights reserved. PII: S0167-9236(01)00103-8
[32] Peter C. Verhoef * Verhoef@few.eur.nl http://www.few.eur.nl/few/people/verhoef and Bas Donkers 1 Donkers@few.eur.nl,” Predicting customer potential value an application in the insurance industry”, Decision Support Systems, Vol. 32 (2) (2001) pp. 189-199 © 2001 Elsevier Science B.V. All rights reserved. PII: S0167-9236(01)00110-5
[33] Philip K. Chan, Wei Fan, Andreas L. Prodromidis and Salvatore J. Stolfo , “Distributed data mining in credit card fraud detection” , IEEE Expert, Vol. 14 (6) (1999) pp. 67-74 © 1999 IEEE. All rights reserved
[34] R. P. Lippmann, ``An introduction to computing with neural nets,'' IEEE ASSP Mag., vol. 4(2), pp. 4-22, April 1987.
[35] R.Rosenblatt,Principles of Neurodynamics. Spartan Books, New York, 1959.
[36] Rumelhart, D.E., Hinton, G.E. and Williams, R.J.(1986). ”Learning internal representation by errorpropagation, ”in Parallel Distributed Processing Vol1.1,318-362,D.E.Rumelhart and L.McClelland(Eds.), M.I.Press,Cambridge,MA.
[37] S. Grossberg, The Adaptive Brian I: Cognition, Learning, Reinforcement, and Rhythm, and The Adaptive Brain II: Vision, Speech, Language, and Motor Control. Elsevier/North-Holland, Amsterdam, 1986.
[38] Sang C. Park a, Selwyn Piramuthu b and Michael J. Shaw c * m-shaw2@uiuc.edu,” Dynamic rule refinement in knowledge-based data mining systems”, Decision Support Systems, Vol. 31 (2) (2001) pp. 205-222 © 2001 Elsevier Science B.V. All rights reserved. PII: S0167-9236(00)00132-9
[39] Sankar K. Pal, Rajat K. De and Jayanta Basak , “Unsupervised feature evaluation: A neuro-fuzzy approach” , IEEE Transactions on Neural Networks, Vol. 11 (2) (2000) pp. 366-376 © 2000 IEEE. All rights reserved
[40] Shang Ming Zhou a * smzhou@btamail.net.cn and Li Da Xu b, “A new type of recurrent fuzzy neural network for modeling dynamic systems” , Knowledge-based Systems, Vol. 14 (5-6) (2001) pp. 243-251 © 2001 Elsevier Science B.V. All rights reserved. PII: S0950-7051(01)00102-2
[41] Sushmita Mitra and Yoichi Hayashi , “Neuro-fuzzy rule generation: survey in soft computing framework” , IEEE Transactions on Neural Networks, Vol. 11 (3) (2000) pp. 748-768 © May 2000 IEEE. All rights reserved
[42] U. M. Fayyad , G. Piatetsky-Shapiro, and P. Smyth , “From data mining to knowledge discovery in databases”, AI Mag., vol. 17, pp. 37-54, 1996.
[43] U. M. Fayyad , G. Piatetsky-Shapiro, and P. Smyth, ” The KDD Process for Extracting Useful Knowledge from Volumes of Data” CACM 39(11): 27-34 (1996)
[44] U. M. Fayyad, G. Piatetsky-Shapiro, and P. Smyth, "From data mining to knowledge discovery in databases", AI Mag., vol. 17, pp. 37-54, 1996.
[45] U. Reimer * ulrich.reimer@swisslife.ch, A. Margelisch and M. Staudt , “EULE: A Knowledge-Based System to Support Business Processes” , Knowledge-based Systems, Vol. 13 (5) (2000) pp. 261-269 © 2000 Elsevier Science B.V. All rights reserved. PII: S0950-7051(00)00086-1
[46] W. Frawley, G. Piatetsky-Shapiro, C. Matheus, “Knowledge discovery in databases: An overview”, in Knowledge Discovery in Databases, G. Piatetsky-Shapiro and W. Frawley, Eds. Menlo Park, CA: AAAI Press, pp. 1-7, 1991.
[47] W. S. McCulloch and W. Pitts, “A logical calculus of the ideas imminent in nervous activity,” Bulletin of Mathematical Biophysics, vol. 5, pp. 115-133, 1943.
[48] 尹振華，美國股票型共同基金分類型態與績效持續性之研究，國立台灣大學財務金融學研究所碩士論文，民國86。
[49] 王若愚，台灣共同基金績效評估－以個股特徵為基準投資組合之研究，國立台灣科技大學管理技術研究所碩士論文，民87國。
[50] 主計處資料庫.
[51] 江怡慧 網路銀行使用者之行為分析 研究報告 2000.
[52] 周育蔚，利用類神經網路建立台灣股價預測模型，台灣大學商學研究所碩士論文，民國85年。
[53] 張瑋倫 , “資料挖掘與顧客關係管理之研究” , 輔仁大學資管研究所碩士論文 ,
[54] 鄭維恆 , “Using dynamic neural network input item to predict customer’s behavior” , 淡江大學資訊研究所碩士論文 ,
[55] 陳炳宏，「共同基金投資組合績效之研究」，國立政治大學企業管理研究所碩士論文，民國87。
[56] 傳銷世紀 專題企劃 “E時代集客新策略” 1997.
[57] 葉怡成，類神經網路模式應用與實作，儒林圖書，台北，1993
[58] 網際先鋒 網路銀行在電子商務中的重要角色 1999.
[59] 數位行銷 顧客流失分析技術.
[60] 顧客關係管理深度解析 遠擎管理顧問公司顧問群 2001.