A test-bed for intrusion detection systems results post-processing

Abstract

Intrusion detection systems produce alert sets of low quality. Many post-processing methods have been proposed to make alert sets more meaningful to security analysts. Relevant research has to deal with an important task; implementing proposed methods and carrying out required experiments. In this paper we propose a platform which can be used as a test-bed for conducting intrusion detection alerts postprocessing research. All the standard functionality is already implemented for the user, as she has to implement only the core logic of her method. Additionally the platform offer important reuse and evaluation capabilities. Finally we use the platform to implement a previous method of ours, in order to test its usefulness.