With the rapid development of mobile Internet,network fault diagnosis technology has become an important research direction.In mobile network fault diagnosis,due to the limited number of fault samples,it is difficult for the traditional methods to accurately diagnose the fault types.Therefore,the study proposes a mobile network fault diagnosis method that fuses Naive Bayesian Model(NBM)and Graph Convolutional Networks(GCN).Through the fusion of GCN and NBM,the fault diagnosis method is able to extract more fault data for fault identification and diagnosis.The results show that the average fault diagnosis accuracy and the aver-age fault false detection of the model method are 92.18％and 9.13％,respectively;meanwhile,the model method has 75.00％effi-ciency in network fault classification identification and the average time overhead at the beginning of fault identification is 11 s.All the results are better than the comparison algorithms,which indicates that the proposed method can effectively identify the mobile net-work fault types and has high accuracy and robustness.
维普
万方数据