Graph clustering methods and equipment

The present specification embodiments provide a graph clustering method and device. The method includes: querying a hyperedge index based on query parameters to obtain a set of core hyperedges of a hypergraph and a set of similar hyperedges for each core hyperedge; using the core hyperedges as clustering starting points, clustering each similar hyperedge in the set of similar hyperedges corresponding to each core hyperedge to obtain multiple clusters; and classifying hyperedges of the hypergraph that are outside the multiple clusters. According to the technical solutions of the present specification embodiments, the computation time for graph clustering can be reduced, thereby enabling efficient structural clustering of large-scale hypergraphs.