Optimization of software vulnerability detection model based on machine learning algorithm
The machine-learning model proposed in the paper has improved performance (efficiency, accuracy, and false-negative rate) for identifying vulnerabilities compared to conventional methods, and shows much greater flexibility and reliability when analyzing large code bases and different categories of vulnerabilities.