A Prediction Model Of Cloud Security Situation Based On Evolutionary Functional Network

As an IT infrastructure,cloud computing has been widely used in various fields.It is because of its openness and complexity that the security problems are more and more serious.Cloud security situational awareness is a process of intelligent reasoning for the optimal decision-making of cloud security and the optimization of cloud security management.Cloud security concerns the future of cloud computing.Cloud security situational awareness includes cloud security situation recognition,cloud security situation comprehension and cloud security situation prediction.The prediction of cloud security situation is an important part of cloud security situation awareness.The prediction of cloud security situation can realize the dynamic management of cloud network security and prevent large-scale cloud security events.In the process of situation recognition,the difficulty lies in how to comprehensively obtain the cloud security situation elements with uncertainty and fuzzy values.In the process of situation understanding,the emphasis is to reduce the dimension of feature space and computational complexity.In the process of situation prediction,the main research object is to improve the accuracy of prediction model output,reduce the error as much as possible and improve the convergence efficiency.In this case,to carry out study of cloud security situation prediction model setup and its construction has high academic and practical significance.At present,the application of artificial neural network to the research of network security situation is the main research method.In this paper,firstly,according to the fact that the functional network can not only solve the problems that can be solved by the general artificial neural network,but also solve the problems that can be solved by the functional network,the performance of the functional network is better than that of the artificial neural network in many aspects,the functional network is introduced into the application of cloud security situation prediction,and the various aspects of the cloud security situation prediction model,its principle and construction method are studied in depth.The main research contents are as follows:(1)By combining evolutionary algorithm and functional network,the evolutionary functional network approximation model is constructed,which can effectively solve the problem of base function selection and coefficient learning,and make up for the shortage that general artificial neural network can only select specific activation function.(2)Using the random approach algorithm to process and understand the cloud security situation elements,a large-scale random approach algorithm is proposed to reduce the computational complexity of the model in data dimensionality reduction.(3)By constructing the credibility matrix of the uncertain influence relationship of security situation elements,we use the multivariate non-linear regression algorithm to predict the cloud security situation.(4)The three prediction models discussed in this paper are validated and analyzed by data sets,and the cloud security situation prediction model based on evolutionary functional network is proved to have higher accuracy and efficiency by simulation analysis and comparative experiments.