Scholarlink

In today’s era, electronic networks are crucial for various aspects of life and work. Ensuring their continuity and confidentiality is vital, as they are vulnerable to attacks that can disrupt operations or leak sensitive information, so it has become necessary to pay attention to it from attacks and risks that may lead to stop or disable the network. Electronic information or the possibility of leaking valuable information to certain parties for the purpose of exploiting it. Moreover, malicious traffic makes network performance inefficient and troubles users. This study investigates various network attacks using data from the Kaggle repository, focusing on HTTP, TCP, and UDP attacks, as well as typical traffic. An advanced deep learning (DL) technology, convolutional neural network (CNN), is utilized in this work. Our experimental results show that CNNs can be used to train intrusion detection systems (IDSs). Our CNN model achieved an accuracy of 99.9%, with precision, recall, and F1 scores of 0.999. We addressed false negatives using cat swarm optimization, reducing their occurrence to zero. The proposed method in real-world IDS deployment, where both security and operational efficiency are critical.

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

Data Science Techniques to Reduce the Occurrence of False Negatives During Intrusion Detection

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Noor Saud Abd,
  • Kamel Karoui,
  • Wisam Dawood Abdullah,
  • Mustafa Abdmajeed Shihab

摘要

In today’s era, electronic networks are crucial for various aspects of life and work. Ensuring their continuity and confidentiality is vital, as they are vulnerable to attacks that can disrupt operations or leak sensitive information, so it has become necessary to pay attention to it from attacks and risks that may lead to stop or disable the network. Electronic information or the possibility of leaking valuable information to certain parties for the purpose of exploiting it. Moreover, malicious traffic makes network performance inefficient and troubles users. This study investigates various network attacks using data from the Kaggle repository, focusing on HTTP, TCP, and UDP attacks, as well as typical traffic. An advanced deep learning (DL) technology, convolutional neural network (CNN), is utilized in this work. Our experimental results show that CNNs can be used to train intrusion detection systems (IDSs). Our CNN model achieved an accuracy of 99.9%, with precision, recall, and F1 scores of 0.999. We addressed false negatives using cat swarm optimization, reducing their occurrence to zero. The proposed method in real-world IDS deployment, where both security and operational efficiency are critical.