Scholarlink

<p>Serverless computing technology is highly emerged in this decade by improving various cloud features such as cost, latency and scalability. This paradigm shift has streamlined server-side management complexities, optimizing workflows for developers. However, a thorough examination of serverless computing and its associated research reveals potential threat factors that could compromise the integrity of cloud computing environments. To address these concerns, this research introduces a novel threat detection framework based on serverless cloud computing. Leveraging the multiple linear regression method for feature selection and pre-processing, the framework incorporates Artificial Intelligence (AI)-infused Machine Learning (ML) algorithms such as Decision Tree (DT) and Random Forest (RF) for the classification of malicious threats. The experimental results demonstrate superior performance compared to previous approaches, with the proposed framework achieving an accuracy rate exceeding 98.98%. Notably, the entire experimentation is conducted in a real-time setup, showcasing the effectiveness of the proposed threat detection framework in outperforming state-of-the-art alternatives.</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

Intelligent threat detection framework for serverless cloud computing architecture using supervised ML algorithms

  • 期刊论文

  • 发表日期：

    2025年8月13日

查看全文
  • Tulasi Kasuba,
  • S. Saravanan,
  • V. V. S. S. S. Balaram

摘要

Serverless computing technology is highly emerged in this decade by improving various cloud features such as cost, latency and scalability. This paradigm shift has streamlined server-side management complexities, optimizing workflows for developers. However, a thorough examination of serverless computing and its associated research reveals potential threat factors that could compromise the integrity of cloud computing environments. To address these concerns, this research introduces a novel threat detection framework based on serverless cloud computing. Leveraging the multiple linear regression method for feature selection and pre-processing, the framework incorporates Artificial Intelligence (AI)-infused Machine Learning (ML) algorithms such as Decision Tree (DT) and Random Forest (RF) for the classification of malicious threats. The experimental results demonstrate superior performance compared to previous approaches, with the proposed framework achieving an accuracy rate exceeding 98.98%. Notably, the entire experimentation is conducted in a real-time setup, showcasing the effectiveness of the proposed threat detection framework in outperforming state-of-the-art alternatives.