Research On Access Control Technologies In Information Sharing Environment

With the rapid development of internet technology and information technology,human beings have entered a new era of network with information sharing as the main feature.Information can be collected,forwarded,stored and processed almost without obstacles via distributed systems,digital media systems,social networks and other channels of communication.However,the development of information sharing technology poses the risk of privacy information leakage to the information publisher.As a pair of contradictory bodies,large-scale open and unpaid information sharing will inevitably lead to privacy disclosure,while excessive and inappropriate privacy protection measures will increase the difficulty of information sharing.The research on information sharing oriented access control technology can not only protect sensitive information from illegal access in various network applications,but also reduce the risk of privacy disclosure brought by information sharing.At the same time,it can ensure that less sensitive information can be shared to the maximum extent.To support the security requirements of information sharing in information systems,this dissertation integrates the technologies of access control,cross-domain forwarding control and access control effect evaluation,and explores the key access control techniques to ensure secure information sharing.The main contents are as follows:(1)A scalable access control model is proposed by discussing comprehensive environment,temporal,role,relationship and other attributes.Based on the application scenario of multidimensional digital media,this dissertation gives the application of privilege scalable access control model in multidimensional digital media systems.The user-authorization attribute assignment relationship and the authorization property-access authority assignment relationship are defined by extending the components such as user,access authority,authorization attribute,etc.According to the User ID,attribute information,the environment,tense and role,the user-authorization attribute assignment relationship assigns the corresponding authorization attribute to the user.According to the authorization attribute given by the user,the authorization attribute access authority relationship assigns the corresponding access permission to the user,so that the scalable access control model can meet the access control requirements of the resources such as digital media in the multidimensional digital media system.Based on the application scenario of social network,by introducing the resource owner(the resource uploader in social network),the relationship between user and user,and combining attributes and relationship types,the mixed access control policy language is defined based on regular expression.Furthermore,an access control policy evaluation algorithm is proposed to enable the scalable access control model to meet the requirements of access control for the text,picture,video and other resources uploaded by users in the social network system.(2)A cross-domain information sharing control mechanism and the security problems of cross domain information sharing are discussed.Taking information forwarding in social network as an example,this dissertation introduces the related concepts of game theory to analyze the incomes of the information forwarder and publisher who adopt different game strategies,and combines the historical data of the forwarding operation to calculate the probability of dishonest forwarding.By comparing with the threshold set by the information publisher,the final decision on whether the forwarding operation is allowed is given,and then a social network forwarding control mechanism based on game theory is proposed,suitable for the cross domain information sharing control mechanism in social networks.(3)To judge the suitability of access control strategy and measure the effect of access control,an evaluation mechanism of access control effect based on Support Vector Machine(SVM)algorithm is proposed,which can effectively evaluate the effect of access control in social networks.Based on the user-user relationship and the description information of the content published by the user,the SVM algorithm is used to implement the access control effect evaluation.Through experiments,it is proved that the mechanism can give an intuitive evaluation value of access control effect and reveal to the user whether the access control strategy has the risk of privacy disclosure,and can effectively guide the use to modify the access control strategy to achieve better privacy protection purposes.