An Information Sharing-Based Multi-Agent Negotiation

Nowadays the electronic commerce market, becoming a market with huge potential, develops very quickly. Due to the explosion type growth of information on the Internet, the increasingly complicated network environment, and the indeterminations of the product supplier all need urgently a kind of new technique, which can actively, intelligently, and dynamically reflect the change of each kind of information, to satisfy a gradually perfect request to the electronic commerce. The software Agent technique brought us to circulate a new mode of the software system in distributed and open environment. The most basic characteristic that the Agent have include: autonomy, interactivity, sociality, face to a target and adapt environment. Applying the Agent technique in the electronic commerce can provide a series of characteristic services, including discovering the resources intelligently, bargaining on the virtual market on the Internet and so on, for the customer and company's house. How to apply the advanced Agent technique in the electronic commerce is a new focus that the recent electronic commerce researchers look for. With the electronic commerce environment complicated and uncontained, the auto negotiation system with excellent function can not only save time but also have no space restriction, so that it carry out convenient bargain then, while in the automatic negotiation according to Agent, the one who represents Agent, exchanges suggestions continuously to reach an agreement which everyone can get benefit from or approve. Therefore, the automatic negotiation problem of the Agent becomes a key with MAS technique. How to improve the ability and efficiency of Agent the exaltation negotiation during the period of handing over with each other is one of the problems that need to resolve in many Agent system research."Information" is a word, which has a long history and plays an important role in negotiating. The information can lead a function first in the business negotiating, can become the basis to choose strategy, and be the key factor that decides success or failure. Particularly in many Agent systems, because of the incompletion of the Agent knowledge, if acquire the related information about the Agent of the other party, such as the hobby, the effect function, prestige, you can be placed in a beneficial position in the negotiating, moreover you can improve the efficiency and reduce the time of negotiation.This paper is based on such a background, research in the Agent automatically negotiations, how to make use of acquaintance between the information-sharing so as to establish the initial belief in improving the efficiency of the negotiations. First presented actuality at home and abroad and the significance of this research are described clearly, also the related theories. By analyzing the advantages and disadvantages of the existing model of negotiations, an information-sharing based on the model of multi-Agent automatically negotiations. In this model, a detailed description of the acquaintance mechanism based on automatic negotiation is put forwarded, and the exchange value of the interaction between the agent a detailed description of the Agent focused on the relationship between changes in the value of. Then, in this paper, a sharing of information on the analysis of algorithms, through the method of negotiations that can forecast the opponents retains value.Then, on the strategy to do a simple test simulation experiments show that the auto-negotiation has made good results. Finally, there are Work Summary and Outlook.