Local Clustering on Complex Graphs and Complex Hypergraphs
Researchers have developed new algorithms, GeneralACL and HyperACL, to improve local clustering on complex graphs and hypergraphs. These algorithms extend the classic Andersen-Chung-Lang (ACL) method to handle weighted, directed, and self-looped graphs, as well as hypergraphs with edge-dependent vertex weights. The new methods are proven to find quadratically optimal clusters in terms of conductance under mild conditions, with experimental validation provided. AI