02050nas|a2200241 i 450000500170000000800410001703500150005810000230007324500630009626000520015930000100021149000270022150001400024850200580038852010150044665000240146170000380148570000480152370000380157171000750160981000590168485600650174320260817161836.0120500s1999    th  uzm rtt  00|  a1eng d  a.b117879090 aSiriwan Apisiridej12aAn agent-mediated negotiation model in electronic commerce  aBangkok :bAsian Institute of Technology,c1999  a93 p.1 aThesis ;vno. IM-99-14  aA thesis submitted in partial fulfillment of the requirements for the degree of Master of  Engineering, School of Advanced Technologies  aThesis (M.Eng.) - Asian Institute of Technology, 1999  aThis study is about the paradigms of negotiation in the Electronic Commerce (EC), in  order to support existing business practices as well as for the new ones over the Internet. Our  work explicitly models negotiation as a multi-agent system that can handle multi-issue  negotiation and use Bayes rule to update belief as the underlying learning mechanism. Besides  considering the trends of counterparts' strategies by using Bayes rule, agents can adapt  learning strategies by considering their environment during a negotiation process via a  Blackboard, which is a globally accessible database-like structure. A hypothetical example of  Bank Loan Negotiation is considered to illustrate the negotiation process for determining the  acceptable amount of loan and interest rate. Production rules are used to represent knowledge  in the analysis and evaluation phase of negotiation. The simulation results show that the  learning strategy with Bayes rule yields better results than the conventional strategies.10aElectronic commerce1 aSadananda, Ramakoti,eChairperson1 aBatanov, Dentcho N.,eExamination Committee1 aYulu, Qi, eExamination committee2 aAsian Institute of Technology Partial Scholarship, eScholarship donor2 aAsian Institute of Technology.tThesis ;vno. IM-99-14  3Full-Textuhttp://203.159.5.9/ait-thesis/detail.php?q=B12344