分散環境上でのマルチエージェントシステムの実現に関する一考察

概要

論文の詳細を見る
In modern societies, information is becoming very valuable and locating then when required constitutes a very hot issue. However, this task put a high load on the user's due to the explosion in the amount of electronic information dispersed over a world-wide large scale information network-like Internet-where traditional retrieval techniques are insufficient. Some agent-based solutions were proposed, but many draw-backs remains still unsolved. In this paper, we introduce a completely agent-based approach for Information Gathering, where several agents come together ; we explain how this "community of agents"-by cooperation and negotiation-decrease user load and gain efficiency in the retrieval process. The proposed multiagent-based architecture has the following merits : (a)sharability : many users can share these expert agents ; (b)complexity hiding : often IG involves quite complex coordination of many different agents. Having the user interact only with his User Agent hides the underlying complexity, (c)modularity and reusabilily : these software agents-pieces of code-can be copied from one user to another with little modification or adaptation to take into account particular user's preferences, (d)flexibility : these software agents can interact in new configurations "on-demand" depending on the information requirements of a particular task.
一般社団法人情報処理学会の論文
1996-03-06