DeKeDVer: A deep learning-based multi-type software vulnerability classification framework using vulnerability description and source code.
Yukun DongYeer TangXiaotong ChengYufei YangPublished in: Inf. Softw. Technol. (2023)
Keyphrases
- source code
- deep learning
- software systems
- open source
- software maintenance
- software projects
- multi type
- software evolution
- high level
- machine learning
- free software
- program understanding
- software repositories
- unsupervised learning
- decision trees
- bug localization
- class labels
- probabilistic model
- feature vectors
- training set
- case study
- source files