Given the increasing existing amount of information techniques for approximate information filtering are necessary. The Minerva Approximate Publish/Subscribe (MAPS) system is designed to support approximate information filtering in a peer-to-peer environment. In the MAPS system a subscriber uses a ranking function to assign a score to each publisher.
The ranking function takes into consideration the document resources of the publishers and their predicted behavior to assign a score to each of them.
We added a price dimension for the information in MAPS and we created an agent based model. The agents use a linear combination between the price of the information and the ranking function to decide which are the best publishers to subscribe to. The purpose of the model is to understand the best practices for publisher and subscribers.