基于多Agent技术的网络科技论文推荐系统

In order to solve the problem that users who do research on interdisciplinary subject cannot gain excellent paper resources, characteristics of the research group is modeled. The relations between different groups are considered, and a new paper resource recommendation system is proposed based on the multi-agent technology. User agent, group agent and recommendation agent are designed to analyze the research qualities of groups. First, similar groups will be found through the interactions of different agents. Second, outstanding groups will be selected from similar groups. Finally, excellent paper resources, shared by outstanding groups, will be recommended. It is proved that the proposed system is useful to the users who do research on interdisciplinary subject in the network.