Software Vulnerability Detection Using Informed Code Graph Pruning

Abstract

pruning methods that can be used to reduce graph size to manageable levels by removing information irrelevant to vulnerabilities, while preserving relevant information. We present “Semantic-enhanced Code Embedding for Vulnerability Detection” (SCEVD), a deep learning model for vulnerability detection that seeks to fill these gaps by using more detailed information about code semantics to select vulnerability-relevant features from code graphs. We propose several heuristic-based pruning methods, implement them as part of SCEVD, and conduct experiments to verify their effectiveness. Our heuristic-based pruning improves on vulnerability detection results by up to 12% over the baseline pruning method.

Keywords

• Code representation
• deep learning
• source code semantics
• vulnerability detection