Department of Information Technology

Abstract: Database security models typically employ information hiding together with polyinstantiation which helps create fictitious and misleading versions of confidential information.Although strict security-control is maintained, a high price is paid in terms of user convenience and data accuracy.Authorized users are often prevented from accessing necessary information which may not compromise security.Moreover, users accessing and unknowingly making decisions based on the misleading information used as cover stories can lead to disastrous consequences. In this paper we employ fuzzy sets in a multilevel model for general purpose database security.Sensitive information in database relations is meaningfully clouded by fuzzy sets; this is accomplished by broadening the possibility distributions constraining the values of sensitive attributes. The technique promotes the use of data and also maintains database security.Clouding with fuzzy sets is the middle ground between information release and information hiding/falsification.It nicely supplements the two security techniques and helps strike the right balance between user convenience and database security.