Publication
Generalizing Local Pattern Mining on Attributed Graphs using Simplicial Complex Abstraction
Mandala von Westenholz; Mika Patzelt; Tim Römer; Martin Atzmueller
In: Hocine Cherifi; Murat Donduran; Luis M. Rocha; Chantal Cherifi; Onur Varol (Hrsg.). Complex Networks & Their Applications XIII - Proceedings of The Thirteenth International Conference on Complex Networks and Their Applications: COMPLEX NETWORKS 2024 - Volume 1. International Conference on Complex Networks and their Applications (COMPLEX NETWORKS-2024), December 10-12, Istanbul, Turkey, Pages 298-311, Studies in Computational Intelligence (SCI), Vol. 1187, Springer, 2025.
Abstract
Local pattern mining on attributed graphs aims to detect subsets of vertices that are induced by patterns composed of specific sets of attributes. This paper generalizes such graph-based approaches to simplicial complexes building on the MinerLSD algorithm for efficient local pattern mining. We present according generalizations of the graph-based closed pattern case to simplicial complexes, as well as a generalization of the Modularity for simplicial complexes for creating such a higher-order pattern mining approach. We demonstrate the advantages of the proposed strategy via experimentation using several datasets.
