Machine Learning-Based Source Code Security Flaw Prediction Method and System

Abstract

This invention relates to the field of code security detection technology, and in particular to a method and system for predicting source code security vulnerabilities based on machine learning. The method includes generating an alert dataset; extracting features from each alert in the dataset to obtain an alert fingerprint for each alert; performing multi-level matching of each alert fingerprint with a preset remediation context fingerprint database to obtain a matching result and matching confidence level for each alert; automatically classifying newly generated alerts based on preset classification rules, according to the matching result and matching confidence level, to obtain classification results; outputting the classification results and corresponding matching evidence, and performing adaptive optimization. This invention can significantly improve the efficiency and accuracy of security engineers in handling source code security vulnerabilities.

Images