KNN Lifting (Graph to Hypergraph) - geometric-intelligence/TopoBench GitHub Wiki

For each node in the graph, the method finds the k nearest nodes by using the Euclidean distance between the vectors of features. The set of k nodes found is considered as an hyperedge. The proces is repeated for all nodes in the graph.

Reference: Hypergraph Learning: Methods and Practices