26-01-2012, 02:12 PM
GUILTY AGENT DETECTION SYSTEM
[attachment=16648]
Proposed System
In this project, developing a model for assessing the “guilt” of agents.
Data allocation strategies are used for distributing objects to agents, in a way that improves our chances of identifying a leaker.
It is possible to assess the likelihood that an agent is responsible for a leak.
Adding of “Fake” objects to the distributed set. Such objects do not correspond to real entities but appear realistic to agents.
The fake objects acts as type of watermark for entire set, without modifying data.
Guilty Agents:
Suppose that after giving objects to agents, the distributor discovers that a set S ⊆ T has leaked.
Say one of the objects in S represents a customer X. Perhaps X is also a customer of some other company, and that company provided the data to the target.
Our goal is to estimate the likelihood that the leaked data came from the agents as opposed to other sources.