Comparing local search with respect to genetic evolution to detect intrusions in computer networks

摘要: The detection of intrusions over computer networks (i.e., network access by non-authorized users) can be cast to the task detecting anomalous patterns traffic. In this case, models normal traffic have determined and compared against current Data mining systems based on genetic algorithms contribute powerful search techniques for acquisition from large amount data made available audit tools. We compare acquired a system distributed algorithm with ones greedy heuristics. Also we provide an empirical proof that representation change result in significant increase classification performances models. Network Information Exploration Shootout project 1998 DARPA Intrusion Detection Evaluation been chosen as experimental testbed.