GCN DEVICE AND METHOD FOR LEARNING ASSEMBLY CODE AND DETECTING SOFTWARE WEAKNESS BASED ON GRAPH CONVOLUTION NETWORK

摘要

Disclosed is a learning apparatus for generating a detection model for detecting the presence or absence of security weaknesses in assembly code. The learning apparatus corresponds to an assembly code constituting the learning data and generates a CFG (Control Flow Graph) composed of nodes and edges, and each of the nodes is pre-sized. and a vectorization unit that converts ? into a predetermined corresponding matrix, and a learning unit that generates a detection model by learning a learning algorithm using a plurality of corresponding matrices each corresponding to one of the nodes and edge information.