社区检测:任何将一个人分配到多个社区的算法?

Community detection: Any algorithm that assigns a person to multiple communities?

我觉得在某些情况下,将一个人分配到多个社区是合理的。例如,在科学领域,一位教授可能在多个研究领域工作。是否有将一个人分配到多个社区或 "soft-labeling approach" 的算法?(例如,与属于任何检测到的社区的人相关的正概率)

谢谢

有很多。术语是“重叠社区检测”。

看看这篇论文,例如,您可以在其中找到 14 种此类算法的比较:

Overlapping Community Detection in Networks: the State of the Art and Comparative Study [Xie, Kelley, Szymanski - 2011]

BIGCLAM, an algorithm presented by Yang and Leskovec is an interesting one, as it scales very well. It is available as part of SNAP.