Scholarlink

A “software-defined network” (SDN) is a type of network architecture used for digitally building and creating hardware components. The network connection’s settings are modifiable at any time. Because the link in a typical network is fixed, dynamic change is not conceivable. Even with SDN being a great tactic, DDoS attacks can still happen. The DDoS attack is putting the internet at risk. The machine learning algorithm can be used to thwart DDoS attacks. A DDoS assault occurs when several systems work together to simultaneously attack a certain host. at SDN, devices at the infrastructure layer are managed by software from the control layer, that sits between the application and network layers. In this post, we suggest a machine learning technique that uses the XGBoost, Adaboost, catboost and RF models to identify fraudulent traffic. The outcomes of our experiment demonstrate the superior detection rates and accuracy of the Random Forest, CatBoost, and XGBoost algorithms.

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

Network Intrusion Attack Identification Relying on Machine Learning for Improving SDN Security

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Tabeen Fatima,
  • K. Ruben Raju,
  • Maddela Parameswar,
  • B. Revathi

摘要

A “software-defined network” (SDN) is a type of network architecture used for digitally building and creating hardware components. The network connection’s settings are modifiable at any time. Because the link in a typical network is fixed, dynamic change is not conceivable. Even with SDN being a great tactic, DDoS attacks can still happen. The DDoS attack is putting the internet at risk. The machine learning algorithm can be used to thwart DDoS attacks. A DDoS assault occurs when several systems work together to simultaneously attack a certain host. at SDN, devices at the infrastructure layer are managed by software from the control layer, that sits between the application and network layers. In this post, we suggest a machine learning technique that uses the XGBoost, Adaboost, catboost and RF models to identify fraudulent traffic. The outcomes of our experiment demonstrate the superior detection rates and accuracy of the Random Forest, CatBoost, and XGBoost algorithms.