Scholarlink

<p>Crypto-ransomware attacks are an escalating global threat, causing significant financial and reputational damage to businesses and individuals. Traditional detection mechanisms, such as signature-based methods, often fail to detect sophisticated ransomware variants that employ advanced evasion tactics. To address these critical shortcomings, this paper introduces RGV2, a next-generation ransomware detection framework that bridges this gap by integrating static and dynamic analysis through Event Tracing for Windows (ETW) logs. RGV2 uses FastText embeddings to extract semantic representations of encoded I/O patterns and leverages deep learning models, such as Conv1D, to classify ransomware based on file I/O behavior. Experimental evaluations show that RGV2 achieves a detection accuracy of 99.64% while maintaining an exceptionally low false-positive rate. These results highlight RGV2’s potential as a scalable and real-time protection mechanism with significant societal impact by ensuring early detection and mitigation of ransomware attacks in real-world environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

RGV2: A Framework for crypto-ransomware detection using FastText and neural networks

  • 期刊论文

  • 发表日期：

    2025年12月26日

查看全文
  • M. Adnan Alvi,
  • Zunera Jalil,
  • Mohammad Imran,
  • Mubbashir Ayub

摘要

Crypto-ransomware attacks are an escalating global threat, causing significant financial and reputational damage to businesses and individuals. Traditional detection mechanisms, such as signature-based methods, often fail to detect sophisticated ransomware variants that employ advanced evasion tactics. To address these critical shortcomings, this paper introduces RGV2, a next-generation ransomware detection framework that bridges this gap by integrating static and dynamic analysis through Event Tracing for Windows (ETW) logs. RGV2 uses FastText embeddings to extract semantic representations of encoded I/O patterns and leverages deep learning models, such as Conv1D, to classify ransomware based on file I/O behavior. Experimental evaluations show that RGV2 achieves a detection accuracy of 99.64% while maintaining an exceptionally low false-positive rate. These results highlight RGV2’s potential as a scalable and real-time protection mechanism with significant societal impact by ensuring early detection and mitigation of ransomware attacks in real-world environments.