Feature extraction and classification algorithm of open source Trojan family variants based on graph network

With the popularity of the Internet and the rapid development of information technology, the problem of computer security has become increasingly prominent, especially the detection and classification of Trojan family variants has become one of the important research topics in the field of network security. Aiming at this problem, this paper proposes a feature extraction and classification algorithm of open source Trojan family variants based on graph network. Firstly, this paper represents the Trojan horse sample as a graph structure, and uses graph network to extract the features of the sample graph, and transforms the information of nodes and edges into feature vectors. Then a classification model based on deep learning is designed, and the extracted features are input into the model for classification prediction. In the experiment, the proposed algorithm is verified and evaluated by using the open Trojan sample data set. The results show that the algorithm has achieved satisfactory results in classification accuracy and stability. In addition, this paper also analyzes the influence of different experimental settings on the performance of the algorithm, and discusses the limitations and improvement space of the algorithm. To sum up, the feature extraction and classification algorithm of open source Trojan family variants based on graph network proposed in this paper has certain application prospects and research significance, and can provide reference for related research in the field of Trojan detection and classification.