Scholarlink

This study introduces a groundbreaking intrusion detection system (IDS) aimed at enhancing Internet security measures in light of the ongoing rise in cyberthreats and the dynamic nature of online security concerns. In particular, a parallel convolutional neural network-long short-term memory (CNN-LSTM) model for reliable feature extraction and the accuracy of the CatBoost classifier for precise classification are harnessed by the suggested strategy. This fusion of approaches strengthens our IDS's defenses against potential data breaches and cyberattacks by enabling it to distinguish between legitimate network traffic and malicious activity. Our innovative IDS model, known as the parallel CNN-LSTM with CatBoost, significantly enhances Internet security by utilizing convolutional neural networks, consequently lowering the sensitivity to data leakage and cyberattacks. This comprehensive strategy stands out as a reliable option for protecting vital online services and data in an era characterized by growing digital vulnerabilities.

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

An Intrusion Detection Model Using Parallel CNN-LSTM with CatBoost

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • M. Parvathy,
  • Jeevaa Katiravan,
  • N. Madhi Vathani,
  • Naafiah Fathima

摘要

This study introduces a groundbreaking intrusion detection system (IDS) aimed at enhancing Internet security measures in light of the ongoing rise in cyberthreats and the dynamic nature of online security concerns. In particular, a parallel convolutional neural network-long short-term memory (CNN-LSTM) model for reliable feature extraction and the accuracy of the CatBoost classifier for precise classification are harnessed by the suggested strategy. This fusion of approaches strengthens our IDS's defenses against potential data breaches and cyberattacks by enabling it to distinguish between legitimate network traffic and malicious activity. Our innovative IDS model, known as the parallel CNN-LSTM with CatBoost, significantly enhances Internet security by utilizing convolutional neural networks, consequently lowering the sensitivity to data leakage and cyberattacks. This comprehensive strategy stands out as a reliable option for protecting vital online services and data in an era characterized by growing digital vulnerabilities.