Mining Model Simulation of Advanced Persistent Threat Data in Cloud Computing Environment
Zhang Zhi-hon · Journal of Shijiazhuang University · 2014
In the cloud computing environment,advanced persistent threat data accurate mining can improve security and defense capability of the cloud computing network. Advanced persistent threat data have nonlinear characteristics of disturbance linear extremum,and the traditional processing method is difficult to achieve accuracy for this class of data mining. An advanced persistent threat data mining simulation model is proposed based on extreme perturbation nonlinear feature extraction in cloud computing environment,and the system load operation is evaluated,so that the dynamic task allocation in cloud computing is obtained. Extreme value analysis of advanced persistent threat data characteristics of non linear disturbance is made,perturbed nonlinear characteristics are extracted,and the steady state probabilities of senior continued threat data are calculated,so that pulse response invariant cycle marker on the nonlinear characteristic is obtained,the advanced persistent threat data extreme value perturbation of the nonlinear characteristics is extracted,and the data mining model is constructed. Simulation results show that the algorithm for the continuing threat data has better correct detection probability of 95% above,and that data mining has superior performance. It has good application value in the cloud computing for advanced persistent threat data detection and mining,providing foundation for the network security system construction