arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.05708cs.SI
收藏

离散Agent群体信息网络中的社团结构

Community Structures in Information Networks for a Discrete Agent Population

Peter Marbach

摘要:

社区是社交网络的一个重要特征。本文的目的是提出一个数学模型来研究社交网络中的社团结构。为此,我们考虑社会网络的一个特殊情况,即信息网络。我们假设有一群对获取内容感兴趣的代理。代理在他们感兴趣的内容类型方面有所不同。代理的目标是形成社区,以最大限度地提高其获取和生产内容的效用。我们用这个模型来描述社区的结构,在这种情况下出现的纳什均衡。在本文中提出的工作概括了在文献中获得的连续代理模型的情况下,一个离散的代理人口模型的情况下的结果。我们注意到,一个离散的代理集更准确地反映了现实生活中的信息网络,并需要为了获得更多的见解社区结构,例如在一个社区内的连接(图结构),以及在一个社区内的信息传播。

英文摘要:

Communities are an important feature of social networks. The goal of this paper is to propose a mathematical model to study the community structure in social networks. For this, we consider a particular case of a social network, namely information networks. We assume that there is a population of agents who are interested in obtaining content. Agents differ in the type of content they are interested in. The goal of agents is to form communities in order to maximize their utility for obtaining and producing content. We use this model to characterize the structure of communities that emerge as a Nash equilibrium in this setting. The work presented in this paper generalizes results in the literature that were obtained for the case of a continuous agent model, to the case of a discrete agent population model. We note that a discrete agent set reflects more accurately real-life information networks, and are needed in order to get additional insights into the community structure, such as for example the connectivity (graph structure) within in a community, as well as information dissemination within a community.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑