Review Article
A Survey of Automatic Software Vulnerability Detection, Program Repair, and Defect Prediction Techniques
Table 5
Technical characteristics of deep learning in static code vulnerability detection.
| System/writer | Analysis object | Vulnerability types | Vector techniques | Network | Dataset |
| VulDeePeckke r [21] | Code gadgets | Library/API function | word2vec | BLSTM | NVD SARD |
| SySeVR [22] | Code gadgets | 126 types of vulnerability | Custom algorithm | 6 deep neural networks | NVD SARD |
| CPGVA [23] | Code stream | CWE-78 CWE-90 | word2vec | CNN, RNN (variant net) | SARD |
| Lee et al. [24] | Function | CWE-121 | instruction2vec | Text-CNN | Juliet |
|
|