摘要
为了解决传统的自主访问、强制访问和基于角色的访问控制方法中存在的隐私信息分级准确性低、数据泄露数量多的问题,提出一种基于反向传播(Back propagation,BP)神经网络的用户信息隐私查询访问控制方法。首先收集用户行为数据,计算用户信任值及用户信息隐私度,将用户信息隐私数据分级,最后将用户行为数据、用户信息隐私度值和用户信息隐私查询要素输入到BP神经网络中,依据BP神经网络实现用户信息隐私查询访问控制。仿真结果表明,与传统方法相比,上述访问控制方法大大降低了误判偏差率,隐私信息分级准确性高,用户信息隐私数据泄露数量少,保证了用户信息隐私安全。
Due to low classification accuracy and large amount of data leakage of privacy information in traditional discretionary access control,the mandatory access control and the access control method based on role,this article presented a method of control the user information privacy query access based on back propagation(BP)neural network.Firstly,we collected user behavior data,and then calculated user trust value and user information privacy degree.After that,we graded the user information privacy data.Finally,we inputted user behavior data,user information privacy value and user information privacy query elements into BP neural network.According to BP neural network,we achieved the control of user information privacy query access.Simulation results show that,compared with the traditional method,the proposed method greatly reduces the error rate of misjudgment.The accuracy of privacy information classification is higher,and the amount of user information privacy data leakage is small.Thus,the user information privacy security can be guaranteed.
作者
马相芬
MA Xiang-fen(Puyang Institute of Engineering,Henan University,Henan Puyang 457000,China)
出处
《计算机仿真》
北大核心
2020年第7期341-345,共5页
Computer Simulation
