Computer Science
Reconstructing the evolution history of networked complex systems
J. Wang, Y. Zhang, et al.
Machine learning can extract the historical formation processes of diverse networked systems— from protein-protein interactions to ecological and social networks—revealing interpretable co-evolution features like preferential attachment, community structure, clustering, and degree correlations that previous theories couldn't jointly explain. Intriguingly, even a model slightly better than random on pairwise link order can reliably restore large-network formation history. This research was conducted by Junya Wang, Yi-Jiao Zhang, Cong Xu, Jiaze Li, Jiachen Sun, Jiarong Xie, Ling Feng, Tianshou Zhou, and Yanqing Hu.
Related Publications
Explore these studies to deepen your understanding of the subject.
