Scholarlink

<p>The rapid proliferation of interconnected devices and Internet of Things (IoT) applications has intensified cybersecurity risks for individuals, organizations, and critical infrastructures. Despite advances in encryption and authentication protocols, networks remain vulnerable to sophisticated cyber threats, particularly Distributed Denial of Service (DDoS) attacks. This study investigates the effectiveness of hybrid machine learning (ML) models in distinguishing between benign traffic and two DDoS variants—DDOS-ACK and DDOS-PSH-ACK. Building on recent advances in intrusion detection, we integrate Fuzzy C-Means clustering with classifiers such as Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), and AdaBoost (ADB). Experiments conducted on benchmark datasets evaluate models using accuracy, precision, recall, F1-score, and Matthews Correlation Coefficient (MCC). Results show that the hybrid models outperform standalone classifiers, particularly in handling imbalanced datasets and overlapping class boundaries. These findings underscore the practical viability of clustering-enhanced ML for real-time, attack-specific intrusion detection systems in IoT and cloud-enabled environments.</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

Machine learning-based hybrid technique to enhance cyber-attack perspective

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月16日

查看全文
  • Aun Abbas,
  • Muqaddas Salahuddin,
  • Muhammad Zohaib Khan,
  • Abdullah Ayub Khan,
  • Fahim Uz Zaman,
  • Syed Azeem Inam,
  • Ghadah Aldehim,
  • Tehseen Mazhar,
  • Muhammad Amir Khan

摘要

The rapid proliferation of interconnected devices and Internet of Things (IoT) applications has intensified cybersecurity risks for individuals, organizations, and critical infrastructures. Despite advances in encryption and authentication protocols, networks remain vulnerable to sophisticated cyber threats, particularly Distributed Denial of Service (DDoS) attacks. This study investigates the effectiveness of hybrid machine learning (ML) models in distinguishing between benign traffic and two DDoS variants—DDOS-ACK and DDOS-PSH-ACK. Building on recent advances in intrusion detection, we integrate Fuzzy C-Means clustering with classifiers such as Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), and AdaBoost (ADB). Experiments conducted on benchmark datasets evaluate models using accuracy, precision, recall, F1-score, and Matthews Correlation Coefficient (MCC). Results show that the hybrid models outperform standalone classifiers, particularly in handling imbalanced datasets and overlapping class boundaries. These findings underscore the practical viability of clustering-enhanced ML for real-time, attack-specific intrusion detection systems in IoT and cloud-enabled environments.