Scholarlink

<p>The Dark Web facilitates numerous illicit activities, presenting significant challenges for law enforcement and cybersecurity professionals due to its sophisticated anonymization techniques. This study introduces a novel hybrid deep learning model that combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to classify Dark Web traffic patterns. By leveraging CNN’s ability to extract spatial features from raw packet data and LSTM’s capacity to model temporal dependencies in traffic flows, the architecture achieves more nuanced traffic classification than traditional machine learning approaches. Evaluated on a diverse dataset of Dark Web traffic samples, our model achieves 98.39% classification accuracy, outperforming standalone CNN and LSTM models. These results establish the model’s efficacy as a practical tool for Dark Web traffic monitoring and cybersecurity applications.</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

Efficient Dark Web traffic classification using a hybrid CNN-LSTM model

  • 期刊论文

  • 发表日期：

    2025年2月21日

查看全文
  • Ngaira Mandela,
  • Sonia,
  • Nilay Mistry,
  • Arpita Nagpal

摘要

The Dark Web facilitates numerous illicit activities, presenting significant challenges for law enforcement and cybersecurity professionals due to its sophisticated anonymization techniques. This study introduces a novel hybrid deep learning model that combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to classify Dark Web traffic patterns. By leveraging CNN’s ability to extract spatial features from raw packet data and LSTM’s capacity to model temporal dependencies in traffic flows, the architecture achieves more nuanced traffic classification than traditional machine learning approaches. Evaluated on a diverse dataset of Dark Web traffic samples, our model achieves 98.39% classification accuracy, outperforming standalone CNN and LSTM models. These results establish the model’s efficacy as a practical tool for Dark Web traffic monitoring and cybersecurity applications.