Departamento de Sistemas Complejos, Instituto de Física, Universidad Nacional Autónoma de México (UNAM), Coyoacán, CDMX, México.
Instituto de Investigaciones en Matemáticas Aplicadas y en Sistemas, Universidad Nacional Autónoma de México (UNAM), Coyoacán, CDMX, México.
PLoS One. 2024 Feb 27;19(2):e0298791. doi: 10.1371/journal.pone.0298791. eCollection 2024.
Although friendship networks have been extensively studied, few models and studies are available to understand the reciprocity of friendship and foes. Here a model is presented to explain the directed friendship and foes network formation observed in experiments of Mexican and Hungarian schools. Within the presented model, each agent has a private opinion and a public one that shares to the group. There are two kinds of interactions between agents. The first kind represent interactions with the neighbors while the other represents the attitude of an agent to the overall public available information. Links between agents evolve as a combination of the public and private information available. Friendship is defined using a fitness function according to the strength of the agent's bonds, clustering coefficient, betweenness centrality and degree. Enmity is defined as very negative links. The model allows us to reproduce the distribution of mentions for friends and foes observed in the experiments, as well as the topology of the directed networks.
尽管友谊网络已经得到了广泛的研究，但很少有模型和研究可以用来理解友谊和敌人的互惠关系。在这里，提出了一个模型来解释在墨西哥和匈牙利学校的实验中观察到的有向友谊和敌人网络的形成。在所提出的模型中，每个代理都有一个私人意见和一个公开意见，公开意见与群体共享。代理之间存在两种类型的交互。第一种代表与邻居的交互，而另一种代表代理对整体公共可用信息的态度。代理之间的链接作为可用公共和私人信息的组合进行演变。友谊是根据代理的联系强度、聚类系数、介数中心度和度使用适合度函数定义的。敌意被定义为非常负面的联系。该模型允许我们再现实验中观察到的朋友和敌人提及的分布，以及有向网络的拓扑结构。
