为刻画真实数据集的多维关系,以综合视角研究复杂的现实系统,学者将研究重点逐渐从复杂网络转向超网络的研究.超网络分为两类,基于网络的超网络和基于超图的超网络.基于超图的超网络,结构简单,更适合刻画现实世界的多元结构,这种独特优势使其获得越来越多的重视.本文综述并讨论这类超网络的研究进展.首先,界定超网络的概念;其次,归纳用于刻画超网络结构的静态拓扑指标;再次,总结刻画超网络动力学过程的演化模型;并梳理其主要应用领域.最后展望超网络的若干发展方向,旨在推进超网络领域的理论和应用研究.
To depict the multiple characteristics of real systems, it is necessary to study complex systems in an integrated perspective. Some scholars focus on networks which "above and beyond the existing network". This kind of networks can be divided into two categories, including supernetworks and hypernetworks. Hypernetworks, with a simple structure, are more suitable for depicting the multi-structure of the real world. Thus they have gained more and more attentions from scholars. This article reviewed the researches on hypernetworks. First, the concept is given. Secondly, topological characteristics used to depict the structure of hypernetworks are summarized. Thirdly, evolving models are presented. Then the applications are demonstrated. Finally, possible directions for future researches are identified, aiming at promoting both the theoretical and applied researches in hypernetworks.