Hypergraph Artificial Benchmark for Community Detection (h-ABCD)
This provides a synthetic benchmark for tuning hypergraph community detection algorithms, but it is incremental as it extends an existing graph model to hypergraphs.
The authors introduced h-ABCD, a hypergraph counterpart to the ABCD model, which generates random hypergraphs with power-law distributions for community sizes and degrees, allowing control over noise and homogeneity of hyperedges within communities.
The Artificial Benchmark for Community Detection (ABCD) graph is a recently introduced random graph model with community structure and power-law distribution for both degrees and community sizes. The model generates graphs with similar properties as the well-known LFR one, and its main parameter can be tuned to mimic its counterpart in the LFR model, the mixing parameter. In this paper, we introduce hypergraph counterpart of the ABCD model, h-ABCD, which produces random hypergraph with distributions of ground-truth community sizes and degrees following power-law. As in the original ABCD, the new model h-ABCD can produce hypergraphs with various levels of noise. More importantly, the model is flexible and can mimic any desired level of homogeneity of hyperedges that fall into one community. As a result, it can be used as a suitable, synthetic playground for analyzing and tuning hypergraph community detection algorithms.