With the continuous development of information technology, the amount of complex network data has increased rapidly, and how to effectively mine and identify the patterns has become a key issue. This article aims to build an efficient complex network Data Mining (DM) and pattern recognition model with the help of Artificial Intelligence (AI) technology. Combined with the characteristics of complex networks, the model is built by using the algorithms of Convolutional Neural Network (CNN), Graph Neural Network (GNN) based on attention mechanism, Autoencoder (AE), Generative Adversarial Network (GAN) and Support Vector Machine (SVM) in Deep Learning (DL). Experiments are carried out on data sets in different fields, such as simulating social networks, power transmission networks, and biological gene regulation networks. The results show that this model performs well in the accuracy, recall and F1 value of DM. On the simulated social network data set, the DM accuracy rate is 92%, the recall rate is 90%, and the F1 value is 91%. The accuracy of pattern recognition is 93%, the recall rate is 91%, and the F1 value is 92%. Compared with the traditional methods such as DBSCAN-KNN, Spectral-NB and MLP-All, this model is significantly ahead in all indicators.
Xiaoyu Zhang (Sun,) studied this question.