An approach for predicting multiple-type overflow vulnerabilities based on combination features and a time series neural network algorithm.
Zhangqi ZhengBing ZhangYongshan LiuJiadong RenXuyang ZhaoQian WangPublished in: Comput. Secur. (2022)
Keyphrases
- neural network
- detection algorithm
- learning algorithm
- optimization algorithm
- cost function
- dynamic programming
- computational cost
- multiple types
- feature extraction
- k means
- np hard
- preprocessing
- feature selection
- matching algorithm
- motif discovery
- training algorithm
- associative memory
- convergence rate
- classification method
- false positives
- particle swarm optimization
- probabilistic model