The Negotiation Model Of Multi-agents System Based On What-This/That Learning Mechanism

With the rapid increasing of computer network, the World Wide Web has become an invaluable information resource but the explosion of available information has led up web-searching to a time consuming and complex thing simultaneously. The large number of information sources and their different levels of accessibility, reliability and correlations expounded a complex problem. At the same time, there are complex problems too difficult to solve oh other fields. Accordingly Agent and Multi-Agent systems emerged as the times require. Whereas, as we know. Why DAI (Distributed Artificial Intelligence) technology has been deployed in relatively few applications after it emerged? That is because it lacks a clear and effective model of negotiation among parts of MAS (Multi-Agents System)to solve problem, which specifies how Agents should operate and interact in complex, dynamic and unpredictable environments. And yet, creditworthy of negotiation required the Agent, wl:ich made up MAS in the capacity of learning by oneself, knowledge expressing and knowledge searching, interacting. As a consequence of the experience gained whilst building a number of MAS for all kind of applications, a new effectual model of cooperation has been developed, which could foresight and transact how individuals should behave both when it is progressing satisfactorily and also when it runs into difficulty.Based on profound investigation and analyses of work modules and characteristic of Agent and MAS, the paper expounded a new negotiation model by analyzing and studyingthe completed negotiation models of MAS between its Agents, which tries to works as person. In conclusion, we exerted ourselves to complete the emulation in achieving the negotiation model. In this paper a new idea is to present in MAS, which based on What-This/That learning mechanism instead of traditional Know-How learning mechanism. In regard to the knowledge by learning, we present that the super-dimensional cube, which has the characteristic to integrate various knowledge representation conveniently, is used as the architecture of knowledge base. The speediness algorithm of knowledge access, the best matching cube selection algorithm, is discussed in this paper when the super-dimensional cube is used as the architecture of knowledge base. The time complexity of the algorithm is analyzed, and the parallel algorithm is discussed also. In the end, we designed boldly a three-dimensional framework for Server Agents in the negotiation model on the ground of the efficiency requirement and the environmental requirement. Finally, we used Aglet terrace to lay out a series JAVA programs, many intelligent programs with superior autonomous interoperate, fulfilled some normal and simple tasks in electronic commerce system; and did plenty of experiments, acquired experimental data and experimental results measure up to theoretical assay.