(18.206.187.91) 您好!臺灣時間:2021/05/19 01:15
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:黃信鑫
研究生(外文):Hsin-Hsing Huang
論文名稱:混合式學習方法於案例式推理之研究與應用
論文名稱(外文):Using a Hybrid Learning Method to Improve Case-Based Reasoning
指導教授:劉立頌
指導教授(外文):Alan Liu
學位類別:碩士
校院名稱:國立中正大學
系所名稱:電機工程研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2001
畢業學年度:89
語文別:中文
論文頁數:54
中文關鍵詞:案例類神經階層式離線式線上式
外文關鍵詞:CBRneural networkdecision tree
相關次數:
  • 被引用被引用:0
  • 點閱點閱:185
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
案例式學習是利用過去的經驗來解決目前所遇到的問題。把經驗存成案例放在案例庫中,藉由問題的相似度的測量來擷取最相似於目前問題的案例來解決問題,並且把相似的案例做修改。然而修改的知識來自於對特定領域的充分認知。在本論文中引進機器學習的方法來改善對於案例式推理系統中的修改知識不足瓶頸。在我們的方法中先把案例表達成階層式的架構,根據此架構把案例分成抽象的案例與具體的案例。之後依據案例特性的不同我們選擇不同的機器學習方法去擷取案例修改知識。其中對於具體的案例我們選擇類神經網路的方法來擷取修改知識。對於這個方法我們稱之為離線學習。另外一方面,為了讓系統能夠對於動態環境的適應,我們加入了線上學習的機制讓系統變的應有彈性。我們利用此一架構用於機器人足球模擬器。我們用案例式推理系統來架構每一個球員,其中用階層的方式來表達機器人團隊的策略,用類神經網路來學習機器人的基本技巧,用動態案例的擷取來達成線上學習的功能。

A case-based reasoning (CBR) system uses the past experience to solve new problems. We use the cases to store past experience and we collect these cases to put into the case base. By the similarity measure, a system retrieves the most similar cases from the case base according to the problem description. After the case retrieval, the system adapts the retrieved case with the case adaptation knowledge. But we are not experts in special domain; hence, it is difficult for the case adaptation knowledge acquirement. For this reason, we import the machine learning techniques to improve the case adaptation accuracy in the CBR. In our method, we use a hierarchy method to represent the cases and according to this method, the case is divided into the abstract cases and the concrete cases. Due to different category cases have different property, we choose suitable machine learning methods to acquire the case adaptation knowledge. Such as, we use neural networks to learning the adaptation knowledge for the concrete cases. We call this method as off-line learning. In another aspect, we employ on-line learning to let a CBR system more flexible. Using this architecture applies on the robot soccer simulator. We use CBR to structure a soccer player, which is used a hierarchy method to represent the team strategy, used the neural network to learn the basic skill and used dynamic case retrieval to achieve the on-line learning.

1. Introduction…………………………………………………………………………1
2. Case-based Reasoning and Machine Learning……………………………………..3
2.1Case-Based Reasoning Process………………………………………………..3
2.2Decision Tree and Neural Network……………………………………………5
2.2.1 Decision Tree Presentation………………………………………………..5
2.2.2 Artificial Neural Networks………………………………………………..6
2.2.2.1Description…………………………………………………………….6
2.2.2.2Backprogation…………………………………………………………7
2.3Case Representation……..……………………………………………………8
2.4Case Retrieval………………………………………………………………..10
2.4.1 Case Index……………………………………………………………….11
2.4.2 Feature Identify……………………………………………………….…12
2.4.3 Case Matching…………………………………………………………...12
2.5Case Adaptation……………………………………………………………...15
2.6Hierarchical Problem Solving………………………………………………..16
2.7Hybrid Case-Based Reasoning Systems……………………………………..19
3. Hybrid Architectures of CBR……………………………………………………...22
3.1Motivation……………………………………………………………………22
3.2Off-line Learning……………………………………………………………..23
3.3On-line Learning……………………………………………………………..27
3.4Combining Off-learning and On-line Learning into CBR…………………...33
3.5Discussion……………………………………………………………………34
4. Building a CBR system in Robot Soccer………………………………………….37
4.1Introduction of Robot Soccer Simulator……………………………………..38
4.2Architectures of My System………………………………………………….40
4.2.1 Case Representation in Robot Soccer of Simulator………………………40
4.2.2 Learning Adaptation Knowledge of Low-Level Skill…………………….41
4.2.3 Learning On-line Case Retrieve of High-Level Strategy…………………42
4.2.3.1The Representation of Strategy Case………………………………...42
4.2.3.2The Indexing of the strategy Case……………………………………43
4.2.4Using Our Architecture in the Simulator of Robot Soccer………………...44
5. Conclusion and Future Work………………………………………………………46
Appendix A Adaptation……………………………………………………………...48
Bibliography…………………………………………………………………………50

[Aamodt and Plaza, 1994] Agnar Aamodt and Enric Plaza, “Case-base Reasoning: Foundation Issues, Methodological Variations, and System Approach”, AI Communication, IOS press. Vol. 7, No. 1, 1994, pp. 39-59.
[Arkin, 1999] Ronald C. Arkin “Behavior Based Robotics”, The MIT Press Cambridge, Massachusetts London, England, ISBN 0-262-01165-4, 1999.
[Bento and Costa, 1993] Carlos Bento and Ernesto Costa, “A Similarity Metric for Retrieval of Cases Imperfectly Explained”, EWCBR 1993, volume 837 of Lecture Notes in Artificial Intelligence, pp. 92-105.
[Bergmann and Wilke, 1996] Ralph Bergmann and Wolfgang Wilke, “PARIS: Flexible Plan Adaptation by Abstraction and Refinement”, ECAI, Workshop: Adaptation in CBR, 1996.
ftp://ftpagr.informatik.uni-kl.de/pub/ECAI96-AADAPT-WS/bergmann.ps
[Bergmann and Wilke, 1996a] Ralph Bergmann and Wolfgang Wilke, “On the Role of Abstract in Case-Based Reasoning” EWCBR, 1996, pp. 28-43.
[Bergmann and Wilke, 1998] Ralph Bergmann and Wolfgang Wolfgang, “Towards a New Formal Model of Transformation Adaptation in Case-Based Reasoning”, GWCBR, 1988.
[Branting and Aha, 1995] L.Karl Branting and David W. Aha, “Stratified Case-Based Reasoning: Reusing Hierarchical Problem Solving Episodes”, IJCAI, 1995, pp. 384-390.
[Corten et al, 1999] Emiel Corten, Klaus Dorer, Fredrik Heintz, Kostas Kostiadis, Johan Kummeneje, Helmut Myritz, Itsuki Noda, Jukka Riekki, Patrick Riley, Peter Stone and Tralvex Yeap “Soccerserver Manual”, RoboCup SoccerServer Manual Account, 19999-07-03
http://www.dsv.su.se/~johank/RoboCup/manual/ver5.1release/browsable/main.html
[Cercone et al, 1999] Nick Cercone, Senior Member, IEEE, Aijun An, and Christine Chan, “Rule-Induction and Case-Based Reasoning: Hybrid Architectures Appear Advantageous”, IEEE, 1999, Volume 11 of Knowledge and Data Engineering, pp. 166-174.
ftp://eivind.imm.dtu.dk/pub/rod-Case_based.ps.gz
[Corchardo et al,1998] J.M. Corchardo, B. Lees, C. Fyfe, N. Rees, J. Aiken, “Neuro-Adaptation Method for a case-Based Reasoning System”, Computing and Information Systems Journal, Vol 5, No 1, pp 15-20, 1 febrero, 1998. ISBN: 1352-9404.
[Han et al, 2000] Eui-Hong Han, George Karypis, Vipin Kumar “Text Categorization Using Weight Adjusted K-Nearest Neighbor Classification”,
http://www.msi.umn.edu/general/Reports/2000.html
[Hanney et al, 1995] Kathleen Hanney, Mark T. Keane and Barry Smyth, “What Kind of Adaptation do CBR Systems Need?: A Review of Current Practice”, AAAI, Adaptation Knowledge for Reuse, 1995.
http://www.aic.nrl.navy.mil/~aha/aaai95-fss/papers.html
[Hanney and Keane, 1997] Kathleen Hanney, and Mark T. Keane “The Adaptation Knowledge Bottleneck: How to Ease it by Learning from Cases”, ICCBR, Second International Conference on Case-Based Reasoning, pp. 359-370, 1997.
[Hsu and Ho, 1998] Chien-Chang Hsu and Cheng-Seen Ho, “A Hybrid Case-Based Medical Diagnosis System”, Tools with Artificial Intelligence: Proceedings. Tenth IEEE International Conference, 1998, pp. 359-366.
[Jiang and Tseng, 2000] M. F. Jiang and S. S. Tseng, “Structural Cases Representation in Case-based Expert Systems”, IEEE, Knowledge-Based Intelligent Engineering Systems and Allied Technologies, Vol 1, pp. 432-435, 2000.
[Khalid and Omatu, 1991] Marzuki Khalid and Shigeru Omatu, “Control of Real-Time Processes Using Backpropagation Neural Networks”, Proc. of IEEE IECON'91, 1991.
[Kolodner, 1993] Janet Kolodner, “Case-Based Reasoning”, Morgan Kaufmann Publishers, Inc, ISBN 1-55860-237-2, Chapter 5,1993.
[Leake et al, 1997a] David B. Leake, Andrew, Kinley, and David Wilson, “A Case Study of Case-Based CBR”, International Conference on Case-Based reasoning, 1997, pp. 371-382.
[Leake et al., 1997b] David B. Leake, Andrew, Kinley, and David Wilson, “Case-Based Similarity Assessment: Estimating Adaptability from Experience”, In Proceedings of the 14th AAAI National Conference on Artificial Intelligence (AAAI'97), pages 674—679, AAAI press, 1997.
[Luger and Stubblefiled,1998] George Luger and William Stubblefield, “Artificial Intelligence: Structures and Strategies for Complex Problem Solving”, Addison Wesley Longman, 1998, ISBN 0-805-31196-3, Chapter 8.
[Murray-Smith and Thaker, 1993] Roderick Murray-Smith, Sunil Thaker “Combining Case Based Reasoning With Neural Networks” AAAI, Workshop: AI in Service and Support, 1993.
[Mitchell, 1997] Tom M, Mitchell “Machine Learning”, The McGraw-Hill Companies, Inc, ISBN 0-07-042807-7, Chapter 6.
[Michalski et al, 1998] Ryszard S. Michalski, Ivan Bratko and Miroslav Kubat, “Machine Learning and Data Mining”, JOHN WILEY & SONS LTD, ISBN 0-471-97199-5, 1998.
[Quinlan, 1993] J. Ross Quinlan “C4.5 Programs for Machine Learning”, Morgan Kaufmaaa Publishers, ISBN 1-55860-238-0, 1993.
[Ram et al, 1992] Ashwin Ram, Ronald C.Arkin, Kenneth Moorman and Russell J. Clark, “A Case-Based Method for On-line Selection and Adaptation of Reactive Control Parameters in Autonomous Robotic Systems”, AAAI, AI for Real-Work Autonomous Robots, 1992.
ftp://ftp.cc.gatech.edu/pub/ai/ram/er-92-07.ps.Z
[Reategui et al, 1995] Reategui, E. B., Campbell, J. A., Borghetti, S. “Using a Neural Network to Learn General Knowledge in a Case-Based System”, Case-Based Reasoning - Research and Development: First International Conference, ICCBR,1995, pp. 528-537.
[Richardson and Warren, 1996] Margaret M. Richardson and James R. Warren, “Induced Decision Trees for Case-Based Reasoning”, IEEE, 1996, Intelligent Information Systems, pp. 52-55, 1996.
[Sebag and Schoenauer, 1993] M. Sebag and M. Schoenauer, “A Rule-Based Similarity Measure”, EWCBR 1993, volume 837 of Lecture Notes in Artificial Intelligence, pp 119-131.
[Stone and Veloso, 1998] Peter Stone Manuela and Manuela Veloso, “A Layered Approach to Learning Client Behaviors in the RoboCup Soccer Server”, In Applied Aritificial Intelligence, Vol 12,1998.
http://www.cs.cmu.edu/~pstone/pstone-papers.html
[Tanaka et al., 1992] Toshikazu Tanaka, Masakazu Hattori and Naomichi Sueda, “Use of Multiple Cases in Case-Based Design”, Conference of Artificail Intelligence for Application, 1992, pp. 233-239.
[Tanaka et al, 1992] Toshikazu Tanaka, Masakazu Hattori and Naomichi Sueda, “Using of Multiple Cases in Case-Based Design”, IEEE, Artificial Intelligence for Application, Proceedings of the Eighth Conference, pp. 233-239, 1992.
[Wess et al., 1993] Wess S.,Althoff K., and Derwand G., “Using K-D Tree to Improve the Retrieval Step in Case-Based Reasoning “,EWCBR, volume 837 of Lecture Notes in Artificial Intelligence,1993.
[Wettschereck and Aha, 1995] Dietrich Wettschereck and David W. Aha,” Weighting Features”, ICCBR, In Proceedings of the 1st International Conference, pp. 347-358, 1995.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
1. 黃郁文(民83)。自我效能概念及其與學業成就表現之關係。諮商與輔導,第106期,39-41。
2. 黃炳煌(民82)。二十一世紀教育應培養出怎樣的國民?研習資訊,10卷1期,12-15。
3. 黃德祥(民84)。國中班級經營的困境與突破。教育資料與研究(第六期),教育論壇(三)。
4. 林佩璇(民83)。合作學習在高級職業學校的應用。技術及職業教育,24,21-23。
5. 韋金龍(民85)。國內大專學生對「合作學習」英語教學活動的看法。教育研究資訊,4卷6期,13-26。
6. 林坤燦(民85)。合作學習方案的認識與實施。國教園地, 55/56,17-21。
7. 郭重吉(民81)。從建構主義的觀點探討中小學物理教育的改進。科學發展月刊, 20(5), 548-570。
8. 張淑美(民81)。合作學習的涵義、實施與成效。教育文粹,21,20-25。
9. 張芳全(民86)。教學法的新典範一合作學習。國教月刊, 44卷3 4期,7-16。
10. 高三福(民83)。從自我效能觀點談自信心。中華體育季刊,7卷1期,107-110。
11. 孫志麟(民80)。自我效能的基本概念及其在教育上的應用。教育研究,22,47-54。
12. 鄒浮安(民80)。高雄中學甄試保送入學之數理資優生學習適應狀況之研究。資優教育季刊,39,38-40。
13. 盧富美(民81)。談合作學習及其教學流程。教師之友,33卷4期,22-25。
14. 謝政權(民86)。從國小美勞科合作學習-淺談美勞科的「躲避球比賽」教樣法。國教天地,121,47-51。