Deep Learning for Contextualized NetFlow-Based Network Intrusion Detection: Methods, Data, Evaluation and Deployment

Abdelkader El Mahdaouy, Issam Ait Yahia, Soufiane Oualil, Ismail Berrada · 2026

Read Deep Learning for Contextualized NetFlow-Based Network Intrusion Detection: Methods, Data, Evaluation and Deployment by Abdelkader El Mahdaouy, Issam Ait Yahia, Soufiane Oualil, Ismail Berrada (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-05594.txt.

Description

Network intrusion detectors once read packet contents directly, but encryption now hides payloads, so defenders lean on flow records that sum up each connection. Abdelkader El Mahdaouy, Issam Ait Yahia, Soufiane Oualil, and Ismail Berrada survey deep learning methods that add context back into flow-based detection. Most current detectors judge each flow alone, yet real attacks spread clues across many flows and hosts over minutes or days.

The survey sorts methods along four kinds of context: time patterns, graph relations, extra data modes, and multiple resolutions. It also warns about evaluation flaws that inflate scores, such as time leakage, poor data splits, and narrow data sets. Deployment limits get equal weight, from streaming state and memory growth to latency budgets and model compression. The authors conclude that context helps when attacks leave clear time or relational structure, but gains depend on strict, causal testing.

Catalog record

CREATORAbdelkader El Mahdaouy, Issam Ait Yahia, Soufiane Oualil, Ismail Berrada
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE57 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-05594

Browse the full library · About the Tower of Babel