本研究以人工智慧所提出之記憶學習法為基礎，進而提出以記憶學習法改善多層次代理人系統創研機制。並且以CMP及動力手工具為載具建構出多層次代理人系統工程知識庫及工程專利知識庫學習分類架構，透過學習分類架構可有效的存取資料，以達到知識在利用之效果。本研究將利用UML塑模技術，將多層次代理人創研機制知識傳遞及記錄知識型態表現出來。而創研流程機制中以XML技術將多層次代理人知識有效傳遞。本研究創研流程為專利代理人可透過使用者之關鍵字、特徵字及模糊同義字指定驅動專利搜尋軟體尋找相關專利，提供使用者參考，在瞭解專利後，再透過工程創研代理人找尋專利資料庫及工程知識庫中相關之知識進行設計，將設計方法配合TRIZ的建議發明法則改善，且由新專利提案進行工程知識庫搜尋相關知識，並進行專利技術迴避分析，以產生初步新專利。再進一步以XML技術結合專利分類、專利分析、專利結構相似度比對及專利鑑價系統使資料能夠有效傳遞，將新專利進行迴避分析及價值評估。所分析之結果透過多層次代理人系統所提供代理人學習資料庫儲存，作為未來使用者找尋相關知識時，多層次代理人能直接提供使用者所須之知識之參考。

Based on the ROTE learning method proposed by artificial intelligence, this paper further proposes to use ROTE learning method to improve the invention and development mechanism of the multi-layer agent system. Also, CMP and power-driven hand tool are used as vehicles to build a knowledge tank of multi-layer agent system engineering and the structure of engineering patent knowledge tank learn categorization. Through learning categorization can effectively access data, and improve the efficiency of knowledge reuse, this paper uses UML model technology to present the knowledge delivery and information recording format of the invention and development mechanism of the multi-layer agent. In the invention and development processing mechanism, XML technology effectively delivers the multi-layer agent knowledge. The invention and development processing in this paper is to ensure that the patent agent can drive the patent search software by specifying the key words, characteristic words and fuzzy synonym to search for the related patent, providing the users to undergo design process by means of engineering invention and development agent, patent database and engineering knowledge tank upon understanding the patent. The patent can then be improved by the integration of design method and the TRIZ invention principle, and by the new patent of proposal, the related knowledge is searched, and carry on the patent technology avoids analysis, so as to generate new patent. By combining patent categorization, patent analysis and patent infringement analysis and patent exchange system, XML technology ensures that the information can be delivered effectively, so as to undergo avoids analysis and value appraisal of the new patent. In other word, the analyzed result is saved by the database provided by the multi-layer agent system, ensuring that the multi-layer agent can directly provide the knowledge for the reference of the users upon their request of the related knowledge in the future.

1.車慧中，IP管理關鍵考量，台北，冠亞智財服份有限公司，民國94年。
2.邱世彬，「UML與Petri-net為基礎之多位代理人供應鏈模型分析」，國立高雄第一科技 大學運輸與倉儲營運系碩士論文，民國91年。
3.Fu, S., J. Y. Chung, W. Dietrich, V. Gottemukkala, M. Cohen and S. Chen “A practical approach to web-based Internet EDI,” IEEE, pp.53-58 (1999).
4.Lee, W. B. and H. C. W. Lau., “Multi-Agent Modeling of Dispersed Manufacturing Networks,” Expert Systems with Applications, Vol.16, pp.297-306 (1999).
5.Matthias, K., Intelligent Information Agents: Agent-based information discovery and management on the internet, Springer-Verlag Berlin Heidelberg New York (1998).
6.Chen, H. and V. Dhar., “Online Query Refinement on Information Retrieval Systems: A Process Model of Searcher/System Interaction,” Proc. SIGIR, pp.115-133 (1990)
7.Murugesan, S., “Intelligent Agents on the Interner and Web,” TENCON ’98. IEEE Region 10 International Conference on Global Connectivity in Energy, Computer, Communication and Control, Vol. 1, pp. 97-102 (1998).
8.Yunwen, Y., “Programming with an intelligent agent. Intelligent Systems,” IEEE [see also IEEE Expert], Vol. 18, No. 3, pp.43-47 (2003).
9.Laukkanen, H., H. Helin and M. Lamanen, “Supporting Nomadic Agent-based Applications in the FIPA Agent Architecture,” Proceedings of the first international jointconference on Autonomous agents and multi-agent system, Bologna, Italy, pp.1348-1355 (2002).
10.Clarke, D. W., TRIZ: Through the Eyes of an American TRIZ Specialist-A Study ofIdeality, Contradictions, Resources, Ideation International, Inc. (1997).
11.Kalakota, R., and A. B. Whinston, Frontiers of Electronic Commerce, Addison-Wesley Publishing Company, Inc.(1996).
12.Sycara, K., A. Pannu, M. Williamson and D. Zeng, “Distributed Intelligent Agents,” IEEE Expert, Vol. 11, No. 6, pp. 36-45 (1996).
13.Wooldridge, M. and N. R. Jennings, “Intelligent Agents: Theory and Practice,” The Knowledge Engineering Review, Vol. 10(2), pp. 115-152(1995).
14.Russell, S., and P. Norvig., “Artificial Intelligence: A Modern Approach,” Prentice-Hall(1995).
15.陳昱彰，「個人化網際網路電子購物代理人」，國立交通大學資訊工程研究所碩士論文，民國87年。
16.陳_能，「建構全球資訊網上決策支援系統之研究」，國立中山大學資訊管理研究所碩士論文，民國86年。
17.Merwe, J. and S. H. Solms, “Electronic commerce with secure intelligent trade agents,” Computers & Security, Vol.17, No.5, pp.435-446 (1998).
18.楊逸群，「助理型論文檢索系統之資料庫設計」，國立臺灣大學資訊工程學研究所碩士論文，民國88年。
19.Noda I, Nakashima H, Handa K. “Programming language gaea and its application for multiagent systems,” In : Workshop on Multi-Agent System and Logic Programming, December (1999).
20.Huhns, M. N., “Agent societies. Magnitude and duration,” Internet Computing IEEE, Vol. 6, No. 1, pp. 79-81 (2002).
21.Hendler, J., “Agents and Semantic Web,” IEEE Intelligent systems, Vol. 16, No. 2, pp. 30-37 (2001).
22.Franklin, S. and A. Grasser, Is it an Agent, or just a Program?: A Taxonomy for Autonomous Agents, Proceedings of the Third International Workshop on Agent Theories, Architectures, and Languages, pp. 21-35 (1996).
23.Benatallah, B., Q., Z. Sheng and M. Dumas, “The Self-Serv environment for Web services composition,” Internet Computing, IEEE, Vol. 7, No. 1, pp. 40-48 (2003).
24.Chester, T.M., “Cross-platform integration with XML and SOAP,” IT Professional, Vol. 3, No. 5, pp. 26-34 (2001).
25.Giovanni Caire (TILAB, formerly CSELT), JADE TUTORIALJADE PROGRAMMING FOR BEGINNERS
http://jade.cselt.it/doc/JADEProgramming-Tutorial-for-beginners.pdf, 2003.
26.Fabio Bellifemine, Giovanni Caire, Tiziana Trucco (TILAB, formerly CSELT)Giovanni Rimassa (University of Parma), JADE PROGRAMMER’S GUIDE
http://jade.cselt.it/doc/programmersguide.pdf,2004.
27.W3C, XML, http://www.w3.org/XML, 1998.
28.http://www.cs.nchu.edu.tw/~djchou/xml/xml.htm
29.http://www.sinica.edu.tw/~ctchang/xml/course/xml-qaml.html
30.曾偉育，「智慧型交通模擬系統之物件導向分析與設計」，朝陽科技大學資訊管理系碩士論文，民國93年。
31.邱世彬，「UML與Petri-net為基礎之多位代理人供應鏈模型分析」國立高雄第一科技大學運輸與倉儲營運系碩士論文，民國91年。
32.蕭詠今，「TechOptimizer 訓練教材Version 1.5」，Invention Machine Corporation，May，1999。
33.Ideation International,Inc.「Ideation International FAQs」
http://www.ideationtriz.com/
34.王建榮、林必窕、林慶福，半導體平坦化CMP技術，第10~13頁，台北，全華科技圖書，民國89年。