English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Anomaly-based intrusion detection promises to detect novel or unknown attackson industrial control systems by modeling expected system behavior and raisingcorresponding alarms for any deviations.As manually creating these behavioralmodels is tedious and error-prone, research focuses on machine learning totrain them automatically, achieving detection rates upwards of 99%. However,these approaches are typically trained not only on benign traffic but also onattacks and then evaluated against the same type of attack used for training.Hence, their actual, real-world performance on unknown (not trained on) attacksremains unclear. In turn, the reported near-perfect detection rates of machinelearning-based intrusion detection might create a false sense of security. Toassess this situation and clarify the real potential of machine learning-basedindustrial intrusion detection, we develop an evaluation methodology andexamine multiple approaches from literature for their performance on unknownattacks (excluded from training). Our results highlight an ineffectiveness indetecting unknown attacks, with detection rates dropping to between 3.2% and14.7% for some types of attacks. Moving forward, we derive recommendations forfurther research on machine learning-based approaches to ensure clarity ontheir ability to detect unknown attacks.

A False Sense of Security? Revisiting the State of Machine Learning-Based Industrial Intrusion Detection