Scholarlink

Cloud computing has gained global acceptance, with organisations increasingly relying on cloud services for their daily business operations. However, the rapid propagation of new malicious code variants with zero-day attacks in the cloud causes uncertainty and widespread alarm as attackers’ intentions are often unknown. This paper introduces a safer computing platform or a model that detects malicious codes. Automated selection of optimal security controls for real-time defence is critical within cloud environments. The paper employs the N-gram algorithm for signature extraction, the C4.5 algorithm for creating signature clusters, and a Python programme for selecting optimal security control. Microsoft Azure and Amazon Web Services cloud platform were used to develop the model. The results show significant performance when detecting, classifying, and selecting an optimal security control for defence. The proposed model demonstrates an effective method for addressing malicious code attacks and displays high adaptability.

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

Automating Selection of Security Controls for Cloud Services Using N-gram and C4.5 Algorithms

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Mpho Erick Maimela,
  • Michael Moeti,
  • Joey Jansen van Vuuren

摘要

Cloud computing has gained global acceptance, with organisations increasingly relying on cloud services for their daily business operations. However, the rapid propagation of new malicious code variants with zero-day attacks in the cloud causes uncertainty and widespread alarm as attackers’ intentions are often unknown. This paper introduces a safer computing platform or a model that detects malicious codes. Automated selection of optimal security controls for real-time defence is critical within cloud environments. The paper employs the N-gram algorithm for signature extraction, the C4.5 algorithm for creating signature clusters, and a Python programme for selecting optimal security control. Microsoft Azure and Amazon Web Services cloud platform were used to develop the model. The results show significant performance when detecting, classifying, and selecting an optimal security control for defence. The proposed model demonstrates an effective method for addressing malicious code attacks and displays high adaptability.