Random Risk Early Warning Method for Accessing Massive Data in High-Density Networks
The data sources in high-density networks are diverse and of varying quality,with issues such as data noise,missing data,and errors.The above issues will interfere with the random risk warning process and reduce the accuracy of the warning.Therefore,a random risk warning method for massive data access in high-density network was proposed.This method first monitored the running state of network data in real time.And then,the enhancement method based on nonlinear independent component estimation was adopted to process the network data.Moreover,principal component analysis was combined with linear discriminant analysis to obtain a discriminant principal compo-nent analysis method,thus extracting the characteristics of network data.Furthermore,risk thresholds were used to judge whether abnormal phenomena occurred in network data.Finally,the early warning of random risks in accessing massive data of high-density network was achieved.The experimental results show that the proposed method has good data monitoring effect and high warning accuracy.And the warning time can be controlled to be less than 90ms.
Network data enhancementPrincipal component analysis methodNetwork data monitoringOnline warningRisk threshold
NETL
NSTL
万方数据
