對手喜好預測是最重要的協商支援項目之一，若能預先了解對手喜好，便有機會主導全局。綜觀目前的相關研究，我們發現仍有以下問題有待解決： 首先，當獲知對手喜好後，如何調整己方策略，以達成預設的目標。其次，協商者的喜好在協商中並非一成不變，如何快速追蹤，以免因誤判而造成損失。我們以Faratin等人提出的協商模型為基礎，配合基因演算法進行對手喜好預測，並大幅放寬了前人研究中常見的參數限制。當了解對手喜好後，我們提出多種目標式來反應預測者的長程目標與協商態度，以便在候選集中挑選合適的提案。因此，本研究提出了三種追蹤公式，以快速察覺對手喜好的變動。針對上述方法，我們使用模擬實驗進行效能驗證。實驗結果顯示，結合本研究提出的預測方法與提案策略，預測一方的效用明顯增加；此外，雙方的效用總和與公平性也同時獲得改善。上述結果說明本研究，確實能協助雙方取得較佳合約，有利於長遠合作關係的建立。

Opponent's preferences prediction is one of the most important negotiation supports. If opponent's preferences are known in advance, the negotiators have better chance to obtain a win-win settlement. Even if a number of researches have been proposed for this topic, however, there are still several issues to be resolved. First, if the complete or partial knowledge of opponent’s preferences is known, the negotiator may need to adjust his strategy to achieve preset goals as efficient as possible. Secondly, if the opponent’s preferences are allowed to be changed in negotiations, the negotiator should discover such a change as soon as possible. To this end, we adopt Faratin’s negotiation model to sketch the preference of negotiators, and apply genetic algorithms to predict the opponents’ preferences by using the negotiation history as input. In addition, three objective functions are proposed to mimic various attitude changes of negotiators which affect the selection of the next proposal from the candidate sets. To trace the change of opponents’ preferences, this study proposed different weighted functions to detect the changes as soon as possible. Simulation results show that a negotiator who applies the proposed prediction method can increase his effectiveness significantly. Also, the proposed method can improve the overall utilities and fairness for both sides. The experimental results demonstrate that the proposed prediction technology really helps in achieving a better contract and is conducive to long-term cooperation relationship.

目錄
目錄	III
表目錄	V
圖目錄	VI
第一章	前言	1
第二章	相關理論與技術	6
2.1	效用模型	6
2.2	自動化協商策略	8
2.3	提案表與目標提案的選擇	11
2.4	基因演算法	12
第三章	考量對手喜好變動之協商支援方法	15
3.1	對手喜好預測方法	15
3.2	在候選提案集中的選擇策略	17
I.	以長程目標為考量之選取方式	18
II.	以提案效果為基礎之選取方式	19
3.3	快速追蹤對手的喜好變動	22
第四章	系統實作	25
4.1	對手喜好預測之問題複雜度	25
4.2	基因編碼	26
4.3	系統運作流程	30
第五章	實驗結果與討論	32
5.1	實驗設計	32
5.2	實驗結果之評量指標	33
5.3	各種協商模式下之實驗結果比較	34
5.3.1	整合式與分配式協商的預測實驗結果	34
5.3.2	不同預測目標式的差異	36
5.3.3	應用預測方法於不同讓步計算方式	37
5.4	對手喜好變動下之預測結果	39
第六章	結論與未來展望	42
參考文獻		43
附錄 A：各種喜好變動追蹤模式之預測結果比較表	47
附錄 B：JGAP簡介	50

表目錄
表2-1 常見的議題效用函數	8
表2-2 提案表範例	11
表3-1 雙方Etype值組合所代表的協商意義	21
表4-1 基因編碼之喜好參數表列	26
表4-2 效用模型中的β值	28
表4-3 時間戰略中的β值	28
表4-4 議題共用參數	29
表5-1 應用預測方法於整合式(Integrative)與分配式(Distributive)協商	36
表5-2 選擇提案時使用方式差異	37
表5-3 二種不同的候選提案產生方式之比較	38
表5-4 權重模式對於喜好變動的追蹤效果(整合式→分配式)	40
表5-5 權重模式對於喜好改變時的協商影響(分配式→整合式)	41

圖目錄
圖2-1 基因演算法中的交配	13
圖2-2 基因演算法中的突變	14
圖3-1 協商動線範例	16
圖4-1 利用β值控制效用函數的曲度	27
圖4-2 基因編碼圖示	29
圖4-3 基因編碼時所需的位元數	30
圖4-4 系統流程圖	31

[1]	楊雅卿 (2006)，“電子商務架構下之協商對手喜好預測”， 淡江大學資管系碩士論文。
[2]	鄭勝仁 (2009)，“WebNSS：一套網頁式協商支援系統及其效能驗證”， 淡江大學資管系碩士論文。
[3]	顧承憲 (2005)，“協商中的對手喜好預測”， 淡江大學資管系碩士論文。
[4]	Baek, J. and Kim, C. (2007), “Learning single-issue negotiation strategies using hierarchical clustering method” Expert Systems with Applications, VOL 32, Iss 2, pp. 606-615.
[5]	Blecherman, B. (1999). “Adopting Automated Negotiation ” Technology in Society 21, pp. 167-174.
[6]	Braun, P., et al. (2006) “e-Negotiation systems and software agents： methods, models, and applications ” Intelligent Decision-Making Support Systems： Foundation, Applications, and Challenges, Chapter 15, Springer Verlag.
[7]	Bui, T., Yen, J., Hu, J., & Sankaran, S. (2001), “A Multi-Attribute Negotiation Support System with Market Signaling for Electronic Markets” Group Decision and Negotiation 10, pp. 515-537.
[8]	Carbonneau, R., Kersten, G., and Vahidov, R. (2008), “Predicting opponent’s moves in electronic negotiations using neural networks” Expert Systems with Applications, VOL 34, Iss 2, pp. 1266-1273.
[9]	Chen, Y. M. and Huang, P.-N. (2009), “Agent-based bilateral multi-issue negotiation scheme for e-market transactions” Applied Soft Computing, VOL 9, Iss 3, pp. 1057-1067.
[10]	Chongming, H. (2004), “Predicting agents tactics in automated negotiation” Intelligent Agent Technology, 2004. (IAT 2004). Proceedings. IEEE/WIC/ACM International Conference on, pp. 127-133.
[11]	Darling, T. A., & Mumpower, J. L. (1990), “Modeling Cognitive Influences on the Dynamics of Negotiations,” Proceedings of the 23rd Annual Hawaii International Conf. on System Science, pp. 22-30.
[12]	Faratin, P., Sierra, C., and Jennings, N. R. (1998), “Negotiation Decision Functions for Autonomous Agents” Int. Journal of Robotics and Autonomous Systems, VOL 24, Iss 3 - 4, pp. 159-182.
[13]	Faratin, P, C. Sierra, N. R. Jennings (2002), "Using Similarity Criteria to Make Issue Trade-Offs in Automated Negotiations," Artificial Intelligence, Vol. 142, No. 2, pp. 205-237
[14]	Foroughi, A. (1997), “Minimizing Negotiation Process Losses with Computerized Negotiation Support Systems,” The Journal of Applied Business Research, 14 (4), pp. 15-26.
[15]	Galin, A., Gross, M., and Gosalker G. (2007), “E-negotiation versus face-to-face negotiation what has changed-if anything, Computers in Human Behavior, 23, pp. 787-797.
[16]	Goh, K.-Y., Teo, H.-H., Wu, H., and Wei, K.-K. (2000), “Computer-supported negotiations： an experimental study of bargaining in electronic commerce” Proceedings of the twenty first international conference on Information systems, pp. 104-116.
[17]	Jain, B. A., & Solomon, J. S. (2000), “The Effect of Task Complexity and Conflict Handling Styles on Computer-Supported Negotiations,” Information & Management 37, pp.161-168.
[18]	Kersten, G. E.,& Noronha, S. J. (1999), “WWW-Based Negotiation Support： Design ,Implementation and Use,” Decision Support Systems 25, pp. 135-154.
[19]	Koen V. Hindriks, Catholijn M. Jonker & Dmytro Tykhonov (2009). “The Benefits of Opponent Models in Negotiation,” Web Intelligence and Intelligent Agent Technology, 2009. (WI-IAT 2009). Proceedings. IEEE/WIC/ACM International Conference on, pp. 439-444.
[20]	Krovi, R., Graesser, A. C., and Pracht, W. E. (1999), “Agent behaviors in virtual negotiation environments” Systems, Man, and Cybernetics, Part C： Applications and Reviews, IEEE Transactions on, VOL 29, Iss 1, pp. 15-25.
[21]	Lau, R. Y. K. (2005), “Towards Genetically Optimised Multi-Agent Multi-Issue Negotiations” Proceedings of the Proceedings of the 38th Annual Hawaii International Conference on System Sciences (HICSS'05), Volume 01, pp. 35.3.
[22]	Lee, C.-F. and Chang, P.-L. (2008), “Evaluations of Tactics for Automated Negotiations,” Group Decision and Negotiation, Vol.17, No.6, pp. 515-539.
[23]	Lee Wei-Min & Hsu Chien-Chang(2010), “An intelligent negotiation strategy prediction system,” Machine Learning and Cybernetics (ICMLC), 2010. Proceedings of the Ninth International Conference on, pp. 3225-3229.
[24]	Lim, J. (2003), “A Conceptual Framework on the Adoption of Negotiation Support Systems,” Information and Software Technology 45, pp. 469-477.
[25]	Lopes, F., Mamede, N., Novais, A. & Coelho,H. (2001), “Negotiation Tactics for Autonomous Agents,” Proceedings of the 12th International Workshop on Database and Expert Systems Applications, pp. 708-714.
[26]	Louta, M., Roussaki, I., and Pechlivanos, L. (2008), “An intelligent agent negotiation strategy in the electronic marketplace environment,” European Journal of Operational Research, 187, pp. 1327-1345.
[27]	Matos, N., Sierra, C., and Jennings, N. R. (1998), “Determining Successful Negotiation Strategies： An Evolutionary Approach” 3rd Int. Conf. on Multi-Agent Systems ICMAS-98, pp. 182-189.
[28]	Meffert, Klaus et al.: JGAP - Java Genetic Algorithms and Genetic Programming Package. URL: http://jgap.sf.net
[29]	Mumpower, J. L. (1991), “The Judgment Policies of Negotiators and the Structure of Negotiation Problems,” Management Science 37 (10), pp 1304-1324.
[30]	Oliver, J. R. (1996-1997), “A Machine-Learning Approach to Automated Negotiation and Prospects for Electronic Commerce,” Journal of Management Information Systems 13 (3), pp. 83-112.
[31]	Raiffa, H., J. Richardson, et al. (2002), Negotiation analysis ： the science and art of collaborative decision making. Cambridge, MA, Belknap Press of Harvard University Press.
[32]	Ramchurn S. D., Sierra, C., Godo, L., and Jennings, N. R. (2007), “Negotiating using rewards,” Artificial Intelligence 171, pp. 805-837.
[33]	Rangaswamy, A.& Shell, G. R. (1997). “Using Computers to Realize Joint Gains in Negotiation： Toward an Electronic Bargaining Table,” Management Science 43 (8), pp. 1147-1163.
[34]	Rau, H., Tsai, M., Chen, C., and Shiang, W. (2006), “Learning-based automated negotiation between shipper and forwarder” Computers & Industrial Engineering, VOL 51, Iss 3, pp. 464-481.
[35]	Ren, F., Zhang, M., and Sim., K. M. (2009), “Adaptive conceding strategies for automated trading agents in dynamic, open markets,” Decision Support Systems 46, pp. 704-716.
[36]	Skylogiannis, T., Antonious, G., Bassiliades, N., Governatori, G., and Bikakis A., (2007),  “DR-NEGOTIATE-A system for automated agent negotiation with defeasible logic-based strategies” data & Knowledge Engineering 63, pp. 362-380.
[37]	Su, Y. W., et al. (2001). “An Internet-Based Negotiation Server for E-Commerce,” The VLDB Journal 10, pp. 72-90.
[38]	Wu, B. and Guo, C. (2010), “The research and improvement on the coordinated-negotiation architecture of Web Service Composition based on Agent in the condition of multi-negotiation concurrency” Computer Supported Cooperative Work in Design (CSCWD), 2010 14th International Conference on, pp. 312-317.
[39]	Yen, J., Lee, H. G., & Bui, T. (1996). “Intelligent Clearinghouse： Electronic Marketplace with Computer-Mediated Negotiation Supports,” Proceedings of the 29th Annual Hawaii International Conference on System Sciences, pp. 219-227.