Research Article
Deep Neural Embedding for Software Vulnerability Discovery: Comparison and Optimization
Table 5
The vulnerable functions and nonvulnerable functions are elaborated in this table. The datasets are derived from 12 open-source projects written in C programming language and the Software Assurance Reference Dataset (SARD) project which contains artificially constructed test cases. In the real-world dataset, the vulnerable functions are labeled based on the description of CVE and NVD. The first column lists the name of the dataset, the second column lists the projects, and the last two columns list the number of vulnerable functions and nonvulnerable functions, respectively.
|