Discarding Similar Data with Autonomic Data Killing Framework Based on High-Level Petri Net Rules: An RSS Implementation

This paper describes the evolutions obtained in the autonomic Data Killing framework that was proposed to eliminate undesirable data. The focus now is about discarding similar data. In order to do it, a modeling method is proposed that uses active rules to be applied through High-level Petri nets. Our method focuses in clustering news in groups by its level of similarity, selecting the newest news of the group and discarding the rest. One experiment has been done in order to proof that method is viable.