Research And Implementation Of Access Control Technology For Safe Production Data Based On Dynamic Behavior Of Users

In recent years,Internet technology has been developing rapidly,and everywhere in the society is using Internet technology.The information level and automation level of coal mine safety production is also improving.Many coal enterprises in China are using the Internet technology to build a multi professional and integrated management information system based on safety,transportation,coal mining,geodesy and water prevention and control.But in the construction of safety production information,enterprises can not easily ignore the problem of network security in the enterprise information management system.Especially in the information management of coal mine safety production,the core database of the system will be inestimable to itself if it is invaded by illegal personnel or competitive manufacturers and a foreign organization.Consequence? Therefore,a coal mine enterprise needs to establish an access control model which accords with the safety production standard,so as to improve the safety of enterprise information system.Access control model and technology,as one of the core technologies of information security,have been adopted by many domestic coal enterprises.At present,many coal mining enterprise management systems generally use role based access control model,but with the rapid development of the Internet,the security requirements of the information system are also changing and showing a variety of characteristics.These changes make the traditional model can not meet the requirements of the integrity of the system.Considering the limitations of the RBAC model itself,including the static allocation properties,the user will get the appropriate permissions of the role once the role is obtained and can be used unrestricted.This situation is prone to abuse of privileges,which does not conform to the principle of minimum permissions in access control.In view of the research and analysis of the existing access control models and their respective problems,a trust based access control(TBAC)model is proposed in this paper.The model adds the attribute of trust to the role based access control model,and combines the identity authentication with the behavior authentication.The TBAC model proposed in this paper is built on the basis of user behavior trust attribute,and the research on user behavior is the key to build the model.Therefore,this paper studies the trust problem of user behavior in network.By studying user behavior habits,we can distinguish potential identities of users.In thispaper,the whole process of user behavior authentication,the preprocessing of user behavior data,user behavior authentication based on machine learning,weak trusted user behavior analysis based on game theory,and the trust value of users are obtained by comprehensive machine learning and game analysis,and the model of access control based on trust is proposed.In coal mine information system,we should realize the monitoring and prevention of abnormal behaviors based on trust prevention and real-time monitoring.In this paper,a coal mine information management system is used to obtain experimental data resources for example analysis.Through the experiment and theoretical analysis,this paper shows that this paper can effectively improve the classification resolution of trusted users,weak trusted users and threat users through the use of machine learning technology,and through the addition of user trust attributes,making access to access.The control model is more robust,which lays a solid theoretical foundation for the security of every enterprise's system.In this paper,the user behavior can be analyzed more efficiently with machine learning technology,and then the game analysis is carried out for the weak trusted users.Finally,the user trust attribute is added to the access control model,making the model more dynamic.In this paper,a large number of real user behavior data can be obtained by building a coal mine information management system.In the end,the traditional role based access control model is improved and the user trust property is introduced to make the security of the system as a whole and not only rely on the identity authentication technology to realize the identity.The dual authentication mechanism of the behavior.The main research results of this paper are as follows:(1)a machine learning based user behavior authentication method is proposed.The support vector machine and neural network are used for modeling.Considering the variety of user behavior data sets,the principal component analysis is used to extract the feature before the support vector machine learning modeling,then the LIBSVM toolkit is used,and the parameter selection is carried out through cross validation.(2)using game theory to analyze two weakly trusted users through machine learning.The user behavior is identified by machine learning technology,and the game analysis is carried out on the behavior data between normal user behavior and abnormal user behavior,and the decision is made by analyzing the final income.(3)put forward a trust based access control model suitable for coal mineinformation system.The traditional role based access control model is improved,and the user trust property is introduced,which makes the security of the whole system not only rely on the identity authentication technology,so as to realize the dual authentication mechanism of identity and behavior.