Hypergraph Random Walks, Laplacians, and Clustering
This work addresses clustering for hypergraph-structured data, which is incremental as it builds on existing methods with specific weightings.
The authors tackled hypergraph clustering by proposing a framework based on random walks with edge-dependent vertex weights, which led to higher-quality clusters compared to existing methods in experiments on real-life datasets.
We propose a flexible framework for clustering hypergraph-structured data based on recently proposed random walks utilizing edge-dependent vertex weights. When incorporating edge-dependent vertex weights (EDVW), a weight is associated with each vertex-hyperedge pair, yielding a weighted incidence matrix of the hypergraph. Such weightings have been utilized in term-document representations of text data sets. We explain how random walks with EDVW serve to construct different hypergraph Laplacian matrices, and then develop a suite of clustering methods that use these incidence matrices and Laplacians for hypergraph clustering. Using several data sets from real-life applications, we compare the performance of these clustering algorithms experimentally against a variety of existing hypergraph clustering methods. We show that the proposed methods produce higher-quality clusters and conclude by highlighting avenues for future work.