Hybrid semantics-based vulnerability detection incorporating a Temporal Convolutional Network and Self-attention Mechanism

Keyphrases

• Vulnerability 100%
• Self-attention Mechanism 100%
• Vulnerability Detection 100%
• Detection Model 100%
• Temporal Convolutional Network 100%
• Semantic Bases 100%
• Hybrid Semantics 100%
• Network Attention 100%
• Source Code 50%
• Deep Learning 50%
• Detection Accuracy 50%
• Detection Efficiency 50%
• Software Vulnerability Detection 50%
• Code Structure 50%
• Structural Information 25%
• Publicly Available 25%
• Convolutional Neural Network 25%
• Detection System 25%
• Recurrent Neural Network 25%
• Systems-based 25%
• Time Efficiency 25%
• Structure Feature 25%
• Image Input 25%
• False Positive Rate 25%
• Novelty Detection 25%
• Automatic Detection 25%
• Degree of Stability 25%
• Deep Neural Network 25%
• Code Patterns 25%
• Semantic Information 25%
• Training Base 25%
• First Use 25%
• Multi-feature Fusion 25%
• False Negative Rate 25%
• Feature Fusion 25%
• Software Vulnerability 25%
• Time Overhead 25%
• Desirable Characteristics 25%
• Feature Design 25%
• Fusion Technology 25%
• Manual Features 25%
• Fusion Neural Network 25%

Computer Science

• Self-Attention Mechanism 100%
• Vulnerability Detection 100%
• Temporal Convolutional Network 100%
• Software Vulnerability 50%
• Source Codes 33%
• Detection Accuracy 33%
• Feature Fusion 33%
• Deep Learning 16%
• Convolutional Neural Network 16%
• Time Efficiency 16%
• Recurrent Neural Network 16%
• Automatic Detection 16%
• Deep Neural Network 16%
• Deep Learning Technique 16%
• False Negative 16%
• False Positive Rate 16%