This study proposed a model to give a full-scale simulation for the dynamics of the collaborations in the dblp dataset. It is a distributed model with the capability of simulating large hypergraphs, namely systems with heterogeneously multinary relationship. Its assembly mechanism of hyperedges is driven by Lotka's law and a cooperative game that maximizes benefit-cost ratio for collaborations. The model is built on a circle to express the game, expressing the cost by the distance between nodes. The benefit of coauthoring with a productive researcher or one with many coauthors is expressed by the cumulative degree or hyperdegree of nodes. The model successfully captures the multimodality of collaboration patterns emerged in the dblp dataset, and reproduces the evolutionary trends of collaboration pattern, degree, hyperdegree, clustering, and giant component over thirty years remarkably well. This model has the potential to be extended to understand the complexity of self-organized systems that evolve mainly driven by specific cooperative games, and would be capable of predicting the behavior patterns of system nodes.

中文翻译：

---

分布式超图模型，用于在dblp中对协作进行全面模拟

这项研究提出了一个模型，可以对dblp数据集中的协作动态进行全面模拟。它是一种具有模拟大型超图的能力的分布式模型，即具有异构多元关系的系统。它的hyperedges组装机制是由洛特卡定律和一个合作博弈驱动的，该博弈最大程度地提高了协作的成本效益比。该模型建立在一个圆上来表示游戏，用节点之间的距离表示成本。与生产性研究人员或与许多合作作者合著的好处是通过结点的累积程度或超度来表示的。该模型成功捕获了dblp数据集中出现的协作模式的多模式，并再现了协作模式，程度，超程度，聚类，和30年来的巨大组成部分非常出色。该模型具有扩展潜力，以了解主要由特定合作游戏驱动的自组织系统的复杂性，并且能够预测系统节点的行为模式。