Data Leakage Detection by Using Fake Objects

Abstract

Modern business activities rely on extensive email exchange. Email leakage have became widespread throughout the world, and severe damage has been caused by these leakages it constitutes a problem for organization. We study the following problem: A data distributor has given sensitive data to a set of supposedly trusted agents (third parties).If the data distributed to the third parties is found in a public\private domain then finding the guilty party is a nontrivial task to a distributor. Traditionally, this leakage of data has handled by water marking technique which requires modification of data. If the watermarked copy is found at Some unauthorized site then distributor claim his ownership. To overcome the disadvantage of using watermark, data allocation strategies are used to improve the probability of identifying guilty third parties. The distributor must assess the likelihood that the leaked data come from one or more agents, as opposed to having been gathered from other means. In this project, we implement and analyze a guilt model that detects the agents using allocation strategies without modifying the original data .the guilt agent is one who leaks a portion of distributed data. We propose data 201C;realistic but fake201D; data records to further improve our chances of detecting leakage and identifying the guilty party. And Algorithms implemented using fake objects will improve the distributor chance of detecting the guilt agent. It is observed that by minimizing the sum objective the chance of detecting guilt agents will increase. We also develop a framework for generating fake objects.