Geometric Deep Learning sub-network extraction for Maximum Clique Enumeration

Vincenza Carchiolo,Marco Grassia,Michele Malgeri,Giuseppe Mangioni
DOI: https://doi.org/10.1371/journal.pone.0296185
IF: 3.7
2024-01-16
PLoS ONE
Abstract:The paper presents an algorithm to approach the problem of Maximum Clique Enumeration, a well known NP-hard problem that have several real world applications. The proposed solution, called LGP-MCE , exploits Geometric Deep Learning, a Machine Learning technique on graphs, to filter out nodes that do not belong to maximum cliques and then applies an exact algorithm to the pruned network. To assess the LGP-MCE , we conducted multiple experiments using a substantial dataset of real-world networks, varying in size, density, and other characteristics. We show that LGP-MCE is able to drastically reduce the running time, while retaining all the maximum cliques.
multidisciplinary sciences
What problem does this paper attempt to address?