Android malicious software detection method based on graph convolutional neural network

The invention relates to an Android malicious software detection method based on a graph convolutional network, and the method comprises the steps: obtaining a to-be-detected Android application, carrying out static analysis, constructing a function call graph which comprises API function nodes and an API call relation, carrying out the simplification processing of the function call graph, retaining core API function nodes related to file access and network application, and carrying out the detection of the Android malicious software. Merging other non-core API function nodes to obtain a function abstract graph of the to-be-detected Android application, obtaining an adjacent matrix of the to-be-detected Android application according to the function abstract graph, performing natural language processing on the function abstract graph, obtaining an initial feature vector of the API function nodes in the function abstract graph, obtaining a feature matrix corresponding to the function abstract graph, and detecti