Introduction:

The increasingly integrated Industrial IoT (IIoT) with industrial systems brings benefits such as intelligent analytics, predictive maintenance, and remote monitoring. However, it also exposes industry systems to malware, cyber attacks, and other security risks.

Objectives:

Machine learning techniques shows promising performance in cyber security, including threats detection, vulnerability analysis, risk assessment, etc. This work aims to investigate how machine learning based techniques can be used in enhancing IIoT security and preventing cyber security events against cyber attack, safety threats, and process disruption.

Methods:

A comprehensive survey was conducted to show how machine learning techniques learn detection malicious activities automatically.

Conclusion:

This work reviewed current literature related to machine learning based methods currently being used in IIoT cyber security, and key methods have been presented and compared in terms of their capability and performance against cyber attacks.

中文翻译：

---

机器学习支持工业物联网安全：挑战、趋势和解决方案

介绍：

工业物联网 (IIoT) 与工业系统的日益集成带来了智能分析、预测性维护和远程监控等优势。然而，它也使行业系统面临恶意软件、网络攻击和其他安全风险。

目标：

机器学习技术在网络安全方面显示出良好的性能，包括威胁检测、漏洞分析、风险评估等。这项工作旨在研究如何使用基于机器学习的技术来增强工业物联网安全并防止网络攻击、安全威胁等网络安全事件和流程中断。

方法：

进行了一项全面的调查，以展示机器学习技术如何自动学习检测恶意活动。

结论：

这项工作回顾了与工业物联网网络安全中当前使用的基于机器学习的方法相关的现有文献，并提出了关键方法并对其针对网络攻击的能力和性能进行了比较。