The Discretization Algorithm for Rough Data and Its Application to Intrusion Detection

摘要: The data processed by intrusion detection systems usually is vague, uncertainty, imprecise and incomplete. Rough Set theory one of the best methods to process this kind data. But can only some discrete So with continuous numerical attributes must be discretized before they are used. Some current discretization algorithms classified reviewed in detail. mathematical descriptions problem given means theory. By fusing entropy we propose a simple fast algorithm based on information loss. applied different samples same from KDDcup99 systems. used reduce so as relieve payload experimental results show that proposed sensitive initial for part all condition attributes. dose not compromise effect it improves response performance model remarkably.