AE-Multi-WGAN: A Robust Multi-Attack Detection IDS for Imbalanced Datasets

Keyphrases

• Attack Detection 100%
• Attack Pattern 20%
• Autoencoder 100%
• Class Imbalance 20%
• Complex Data Distribution 20%
• Complex Network Environment 20%
• Computational Load 20%
• Deep Learning Framework 20%
• Deep Learning Model 20%
• Denoising 20%
• Ensemble Deep Learning 20%
• F1 Score 20%
• False Positive Rate 20%
• Feature Compression 20%
• Generative Adversarial Networks 40%
• Imbalanced Data 100%
• Intrusion Detection System 100%
• Low Accuracy 20%
• Malicious Behavior 20%
• Mode Collapse 20%
• Model Complex 20%
• Multi-attack 100%
• Multiple Discriminators 20%
• Multiple Generators 20%
• Network Intrusion Detection System 20%
• Network Scheme 20%
• NSL-KDD 20%
• Precision Score 20%
• Precision-recall 20%
• Recall Score 20%
• Stable Training 20%
• Statistical Techniques 20%
• Still Face 20%
• Synthetic Sample Generation 20%
• Synthetic Samples 20%
• System Method 20%
• System Performance 20%
• Traditional Machine Learning 20%
• Training Difficulty 20%
• Training Instability 20%
• Unseen Attacks 20%
• UNSW-NB15 20%
• Wasserstein GAN 100%
• Wasserstein Generative Adversarial Network 20%

Computer Science

• Autoencoder 100%
• Baseline Method 20%
• Class Imbalance 20%
• Complex Network 20%
• Computational Load 20%
• Data Distribution 20%
• de-noising 20%
• Deep Learning Method 20%
• Deep Learning Model 20%
• Discriminator 20%
• False Positive 20%
• False Positive Rate 20%
• Generative Adversarial Networks 60%
• Intrusion Detection System 100%
• Learning Framework 20%
• Learning System 20%
• Machine Learning 20%
• Malicious Activity 20%
• Network Based Intrusion Detection System 20%
• Network Environment 20%
• Statistical Technique 20%
• Systems Performance 20%