2026-04-17T15:42:41Zhttps://recercat.cat/oai/request

oai:recercat.cat:2117/3854442025-07-24T18:17:02Zcom_2072_1033col_2072_452951

Detection of nework attacks using graph neural networks Cobo Arróniz, Guillermo Universitat Politècnica de Catalunya. Departament d'Arquitectura de Computadors Barlet Ros, Pere Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Telemàtica i xarxes d'ordinadors Neural networks (Computer science) Computer security Artificial intelligence Security AI GNNs Networks Artificial Intelligence Software Network Security Graphs Graph Neural Networks Xarxes neuronals (Informàtica) Seguretat informàtica Intel·ligència artificial The last few years have seen an increasing wave of attacks with serious economic and privacy damages, which evinces the need for accurate Network Intrusion Detection Systems (NIDS). Recent works proposed the use of Machine Learning (ML) techniques for building such systems (e.g., decision trees, neural networks). However, existing ML-based NIDS do not generalize well to other network scenarios and they are barely robust to common adversarial attacks. This TFM will explore the potential of using graph representations of network flows together with Graph Neural Networks for building more robust NIDS that can better generalize to other networks. 2023-01-20 Master thesis https://hdl.handle.net/2117/385444 ETSETB-230.175207 eng S'autoritza la difusió de l'obra mitjançant la llicència Creative Commons o similar 'Reconeixement-NoComercial- SenseObraDerivada' Open Access application/zip application/pdf Universitat Politècnica de Catalunya