Data Leakage is one of the major concerns of the corporate world today. When a company outsources its confidential data to a third party e.g. to a service provider, it may be the case that some of the data are leaked and found in an unauthorized place (e.g., on the web or somebody’s laptop). This would cause a large setback to the company. Thus Data Leakage Detection needs to be done in order to find out the guilty party. In the proposed system an Agent Guilt Model is developed wherein the distributor can assess the likelihood that the leaked data came from one or more agents, as opposed to having been independently gathered by other means. In order to increase the chances of detecting the data leakage, the proposed system also deals with the addition of dummy objects to the real objects.