Scholarlink

The modern digital age has seen an exponential growth in internet penetration, which in turn has increased the probability of a denial-of-service attack. When an evil user launches a Denial of Service (DoS) attack, it’s with the goal of flooding computing and network resources and blocking access to them for normal users. These kind of assaults may happen at any level of the OSI model, from the application layer all the way down to the transport layer. This research focuses on application layer DoS attacks instead of transport and network DoS attacks and attempts to detect Denial of Service using Machine Learning and Neural Network approaches. An updated dataset of Denial of Service attacks divided into many subcategories was used in this experiment. According to the research, the decision tree approach performed well in terms of accuracy and splitting compared to the logistic regression approach.

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

Identifying Application-Level Distributed Denial-of-Service (DDoS) Attacks with Machine Learning (ML)

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Muhammad Atif Yaqub,
  • Muzzamil Mustafa,
  • Saad Hussain Chuhan,
  • Basit Sattar,
  • Muhammad Zulkifl Hasan,
  • Muhammad Zunnurain Hussain,
  • Muhammad Saad Qaiser,
  • Zain ul Abdin,
  • Afshan Bilal,
  • Jibran Ali

摘要

The modern digital age has seen an exponential growth in internet penetration, which in turn has increased the probability of a denial-of-service attack. When an evil user launches a Denial of Service (DoS) attack, it’s with the goal of flooding computing and network resources and blocking access to them for normal users. These kind of assaults may happen at any level of the OSI model, from the application layer all the way down to the transport layer. This research focuses on application layer DoS attacks instead of transport and network DoS attacks and attempts to detect Denial of Service using Machine Learning and Neural Network approaches. An updated dataset of Denial of Service attacks divided into many subcategories was used in this experiment. According to the research, the decision tree approach performed well in terms of accuracy and splitting compared to the logistic regression approach.