Research on Vulnerability Classification Method Based on DMO-PSO-SVM Algorithm

Network vulnerabilities are very important for network security. How to standardize and reasonably classify vulnerabilities is the basic work of current network security related research. At present, various types of vulnerability classification algorithms often require long training time and test time, and are easy to fall into local optimum. Therefore, this paper proposes a new DMO-PSO-SVM algorithm for the classification of vulnerabilities. This algorithm introduces an improved Particle Swarm algorithm to optimize its parameters based on Support Vector Machine. The use of dual mutation operators avoids the occurrence of local optimal conditions, including edge mutation operator and global mutation operator. Experiments show that the method proposed in this paper significantly improves the accuracy of classification of vulnerabilities and reduces the classification time.