Facilitating the interactions among reasoning components by multiple agents

Establishing automatic interaction among reasoning components is a vital issue to exploit Knowledge-Based System. But traditional Conceptual Method can not solve this problem effectively. For this, the primary objective of this paper is to research automatic interactions among the reasoning components without assumptions using multiple Agents. Firstly, a distributed environment oriented to knowledge component is constructed, and an interactive architecture of reasoning components, facilitated by Multi-Agent is proposed. And then, some fundamental components in the distributed environment is designed so that reasoning components can transfer between two interaction patterns, i.e. querying domain knowledge and solving problem cooperatively. Finally, practical procedure of the interactions facilitated by Multi-Agent is presented, and a duple interaction pattern among the reasoning components is established. The result of analysis shows that the approach presented in this paper provides a self-adaptive and more practical solution for researching the interactions among reasoning components.