Research Article

Similarity HyperGraph and Similarity SuperHyperGraph in Medical Science

Abstract

A similarity graph represents items as vertices, with edges connecting pairs whose similarity exceeds a specified threshold or meets k-nearest-neighbor criteria. Applications of similarity graphs have been explored in domains such as medical science. A hypergraph extends this notion by allowing each hyperedge to join any nonempty subset of vertices simultaneously [1–4]. A SuperHyperGraph further introduces a hierarchy by iterating the powerset construction, thereby capturing nested, multi-scale relationships among vertices and edges. In this paper, we investigate Similarity HyperGraphs and Similarity SuperHyperGraphs. These extensions make it possible to study similarity graphs in a more deeply hierarchical context.