Prediction and evaluation of wireless network data transmission security risk based on machine learning

Abstract

The security of wireless network transmission data is an important technical index to ensure the reliable transmission of information in local areas, in this paper, there are a lot of personal privacy in wireless network transmission data, and the consequences of leakage are serious. This paper puts forward the prediction and evaluation of wireless network data transmission security risk based on machine learning, an effective method to solve information leakage and privacy protection uses improved Naive Bayesian kernel estimation (INBK) in machine learning to evaluate wireless network data security and risk level. The results show that the proposed model has lower false positive rate and false positive rate than other methods. In the same type of comparison, as the number of attacking nodes increases, Different algorithms have a certain increase in the false positive rate and the false negative rate. The method proposed in this paper has the advantages of accuracy, the recall rate and F1 algorithm perform well. Four algorithms are on the label U2R, R2L performed poorly, overall, it is over 80%, the overall performance is the best. The risk assessment level shows that the correct rate of the method adopted in this paper is higher than 95% in security risk assessment. Other methods are about 80%, and the worst is only 75%. The overall time consumption of different nodes is 18 ms. The highest average time of other models is 35 ms, and the overall time consumption is more.