利用 NLP 和心理语言学标记分析实现社会工程攻击的实时检测

<jats:p>This article examines the problem of countering social engineering attacks (SEA). To detect them, we propose a hybrid model combining psycholinguistic analysis and NLP methods to identify linguistic markers of manipulation in text. We describe the model's architecture and the concept of integrating it into a real-time system for generating interpretable warnings to the user</jats:p> <jats:p>本文探讨了对抗社会工程攻击 SEA 的问题。为实现其检测 提出了一种结合心理语言学分析与 NLP 方法的混合模型 用于识别文本中的操纵性语言标记。文中描述了该模型的架构以及将其集成为实时运行系统的概念 以向用户生成可解释的预警信息</jats:p>