Impact of Decision-Region Based Classi cation Mining Algorithms on Database Security

One of the challenges facing the computer science community is the development of techniques and tools to discover new and useful information from large collections of data. There are a number of basic issues associated with this challenge and many are still un-resolved. This situation has led to the emergence of a new area of study called \Knowledge Discovery in Databases" (KDD). The recent eeorts of KDD researchers have focused primarily on issues surrounding the individual steps of the discovery process. Those issues that are not directly related to the discovery process have received much less attention. One such issue is the impact of this new technology on database security. In this paper , we investigate issues pertaining to the assessment of the impact of classiication mining on database security. In particular, the security threat presented by a category of classiication mining algorithms that we refer to as decision-region based is analyzed. Providing safeguards against this threat requires, in part, the development of new security policies. Our speciic contributions are the proposal of a set of security policies for use in the context of decision-region based classi-cation mining algorithms along with the speciication and implementation of a security risk measure that allows for the realization of a subset of the proposed policies.