Scholarlink

As the cyber threat landscape advances with intricate malware and persistent cyberattacks, reliable detection mechanisms are becoming increasingly critical. Traditional techniques, such as pattern-based and behavioral analysis, struggle to cope with rapidly evolving malicious threats. While data-driven algorithms provide an alternative, their dependence on manual feature crafting often results in inefficiencies and inaccuracies. This study presents an innovative cybersecurity framework utilizing advanced neural network architectures, focusing on Portable Executable (PE) file inspection. By fine-tuning the parameters of Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN), the solution significantly boosts detection effectiveness and reliability. The primary aim is to mitigate the shortcomings of traditional and machine-based techniques through a refined application of neural computing models. The strategy involves a comparative investigation, highlighting the framework's advantages over existing methodologies. The outcomes illustrate the system's exceptional flexibility, precision, and capability in tackling emerging cyber threats, fostering the development of resilient and adaptable threat detection systems.

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

Improving Malware Identification Abilities Through Deep Learning with Sophisticated Hyperparameter Optimization

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Walid El Mouhtadi,
  • Yassine Maleh,
  • Soufyane Mounir

摘要

As the cyber threat landscape advances with intricate malware and persistent cyberattacks, reliable detection mechanisms are becoming increasingly critical. Traditional techniques, such as pattern-based and behavioral analysis, struggle to cope with rapidly evolving malicious threats. While data-driven algorithms provide an alternative, their dependence on manual feature crafting often results in inefficiencies and inaccuracies. This study presents an innovative cybersecurity framework utilizing advanced neural network architectures, focusing on Portable Executable (PE) file inspection. By fine-tuning the parameters of Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN), the solution significantly boosts detection effectiveness and reliability. The primary aim is to mitigate the shortcomings of traditional and machine-based techniques through a refined application of neural computing models. The strategy involves a comparative investigation, highlighting the framework's advantages over existing methodologies. The outcomes illustrate the system's exceptional flexibility, precision, and capability in tackling emerging cyber threats, fostering the development of resilient and adaptable threat detection systems.