Research On Risk Adaptive Access Control For Big Data Application

Since the emergence of transferable information,people continue to use their ingenuity in order to protect important information.Therefore,cryptography and other related disciplines for the secret transmission of information have been derived.With the Internet of Things and the intelligent interconnection of terminals,more and more information is gathered in the cloud,making data grow exponentially,and with it the rise of big data.Big data security has become an important research topic today.How to judge the risk value in the process of access control and predict the risk in the process of access in the big data environment has become a hot topic in the research of big data access control.This paper studies and analyzes the evaluation principles and evaluation methods in risk adaptive access control,and analyzes the status of attribute-based access control in the context of big data.Comprehensively consider the sub-attributes of the subject,object,work,environment and other attributes in access control to the risk in the access control process,and take the degree of impact on risk as an important analysis factor for establishing a risk assessment index system.Optimize and improve the related knowledge in the areas of analytic hierarchy process,entropy weight method,fuzzy inference and other original risk assessment methods,and propose an improved analytic hierarchy process risk assessment model and adaptive neural fuzzy system risk assessment prediction model,and Perform simulation experiments to verify the effectiveness and accuracy of the risk model.The main work of this article is as follows:(1)Construction of risk assessment index system.Based on an in-depth understanding of risk assessment criteria and risk access control calculation methods,combined with the traditional risk access control assessment model for risk factor analysis and judgment and the establishment of an index system,a comprehensive consideration of the characteristics of the big data application environment,and the construction of big data-based Apply the environmental risk assessment indicator system and analyze and study the risk indicators to establish the final risk influencing factors.(2)Improve and optimize the analytic hierarchy process,and establish a risk assessment model using this method.By analyzing the shortcomings of the analytic hierarchy process in the application of risk assessment,and combining the three-scale construction method to establish an evaluation matrix,the entropy weight method improves the hierarchical weights and constructs a corresponding risk assessment model.In the process of determining hierarchical weights,a judgment matrix is established by inherently assigning each access control attribute,and the historical access records of each attribute are used to construct the weight of the judgment matrix according to the entropy weight method,so that the weights are more objective.When the risk value is finally determined,a fuzzy set is combined with the fuzzy number to reduce subjectivity and provide data for the next risk prediction model.(3)Risk assessment prediction model based on adaptive neural fuzzy system.By studying existing neural networks and fuzzy inference methods,an adaptive neural fuzzy inference system combining neural networks and fuzzy inference systems is established,and corresponding risk assessment and prediction models are constructed through adaptive learning.(4)Simulation.First,establish a risk assessment index system,and then use the risk assessment model of the improved analytic hierarchy process and the risk assessment prediction model of the adaptive neural fuzzy inference system to simulate the risk value in the access control process,and verify the proposed model through experiments.